Hugging Face's logo

eaddario
/
Watt-Tool-8B-GGUF

Text Generation
GGUF
English
quant
experimental
conversational
Model card Files
xet
Community
2
eaddario commited on
Commit
42881fc
·
verified ·
1 Parent(s): 10312a8

Add GGUF internal file structure

Browse files
Files changed (13) hide show
  1. scores/Watt-Tool-8B-F16.md +978 -0
  2. scores/Watt-Tool-8B-IQ3_M.md +982 -0
  3. scores/Watt-Tool-8B-IQ3_S.md +982 -0
  4. scores/Watt-Tool-8B-IQ4_NL.md +982 -0
  5. scores/Watt-Tool-8B-Q3_K_L.md +982 -0
  6. scores/Watt-Tool-8B-Q3_K_M.md +982 -0
  7. scores/Watt-Tool-8B-Q3_K_S.md +982 -0
  8. scores/Watt-Tool-8B-Q4_K_M.md +982 -0
  9. scores/Watt-Tool-8B-Q4_K_S.md +982 -0
  10. scores/Watt-Tool-8B-Q5_K_M.md +982 -0
  11. scores/Watt-Tool-8B-Q5_K_S.md +982 -0
  12. scores/Watt-Tool-8B-Q6_K.md +982 -0
  13. scores/Watt-Tool-8B-Q8_0.md +982 -0
scores/Watt-Tool-8B-F16.md ADDED
@@ -0,0 +1,978 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-F16.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 39 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 36 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | general.file_type | 1 |
38
+ | 28 | UINT32 | 1 | llama.vocab_size | 128256 |
39
+ | 29 | UINT32 | 1 | llama.rope.dimension_count | 128 |
40
+ | 30 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
41
+ | 31 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
42
+ | 32 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
43
+ | 33 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
44
+ | 34 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
47
+ | 37 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
48
+ | 38 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
49
+ | 39 | UINT32 | 1 | general.quantization_version | 2 |
50
+
51
+ ## Tensors Overview ~8B Elements
52
+
53
+ Total number of elements in all tensors: 8030261312 Elements
54
+
55
+ - [Watt-Tool-8B-F16.gguf - GGUF Internal File Dump](#watt-tool-8b-f16gguf---gguf-internal-file-dump)
56
+ - [Key Value Metadata Store](#key-value-metadata-store)
57
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
58
+ - [Tensor Data Offset](#tensor-data-offset)
59
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
60
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
61
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
62
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
63
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
64
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
65
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
66
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
67
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
68
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
69
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
70
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
71
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
72
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
73
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
74
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
75
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
76
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
77
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
78
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
79
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
80
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
81
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
82
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
83
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
84
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
85
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
86
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
87
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
88
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
89
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
90
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
91
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
92
+
93
+ ### Tensor Data Offset
94
+
95
+ This table contains the offset and data segment relative to start of file
96
+
97
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
98
+ |-----:|:--------------------------|-----------------:|-----------------:|
99
+ | 0 | rope_freqs.weight | 0x779520 | 0x100 |
100
+ | 1 | output.weight | 0x779620 | 0x3ea00000 |
101
+ | 2 | token_embd.weight | 0x3f179620 | 0x3ea00000 |
102
+ | 3 | blk.0.attn_norm.weight | 0x7db79620 | 0x4000 |
103
+ | 4 | blk.0.ffn_down.weight | 0x7db7d620 | 0x7000000 |
104
+ | 5 | blk.0.ffn_gate.weight | 0x84b7d620 | 0x7000000 |
105
+ | 6 | blk.0.ffn_up.weight | 0x8bb7d620 | 0x7000000 |
106
+ | 7 | blk.0.ffn_norm.weight | 0x92b7d620 | 0x4000 |
107
+ | 8 | blk.0.attn_k.weight | 0x92b81620 | 0x800000 |
108
+ | 9 | blk.0.attn_output.weight | 0x93381620 | 0x2000000 |
109
+ | 10 | blk.0.attn_q.weight | 0x95381620 | 0x2000000 |
110
+ | 11 | blk.0.attn_v.weight | 0x97381620 | 0x800000 |
111
+ | 12 | blk.1.attn_norm.weight | 0x97b81620 | 0x4000 |
112
+ | 13 | blk.1.ffn_down.weight | 0x97b85620 | 0x7000000 |
113
+ | 14 | blk.1.ffn_gate.weight | 0x9eb85620 | 0x7000000 |
114
+ | 15 | blk.1.ffn_up.weight | 0xa5b85620 | 0x7000000 |
115
+ | 16 | blk.1.ffn_norm.weight | 0xacb85620 | 0x4000 |
116
+ | 17 | blk.1.attn_k.weight | 0xacb89620 | 0x800000 |
117
+ | 18 | blk.1.attn_output.weight | 0xad389620 | 0x2000000 |
118
+ | 19 | blk.1.attn_q.weight | 0xaf389620 | 0x2000000 |
119
+ | 20 | blk.1.attn_v.weight | 0xb1389620 | 0x800000 |
120
+ | 21 | blk.10.attn_norm.weight | 0xb1b89620 | 0x4000 |
121
+ | 22 | blk.10.ffn_down.weight | 0xb1b8d620 | 0x7000000 |
122
+ | 23 | blk.10.ffn_gate.weight | 0xb8b8d620 | 0x7000000 |
123
+ | 24 | blk.10.ffn_up.weight | 0xbfb8d620 | 0x7000000 |
124
+ | 25 | blk.10.ffn_norm.weight | 0xc6b8d620 | 0x4000 |
125
+ | 26 | blk.10.attn_k.weight | 0xc6b91620 | 0x800000 |
126
+ | 27 | blk.10.attn_output.weight | 0xc7391620 | 0x2000000 |
127
+ | 28 | blk.10.attn_q.weight | 0xc9391620 | 0x2000000 |
128
+ | 29 | blk.10.attn_v.weight | 0xcb391620 | 0x800000 |
129
+ | 30 | blk.11.attn_norm.weight | 0xcbb91620 | 0x4000 |
130
+ | 31 | blk.11.ffn_down.weight | 0xcbb95620 | 0x7000000 |
131
+ | 32 | blk.11.ffn_gate.weight | 0xd2b95620 | 0x7000000 |
132
+ | 33 | blk.11.ffn_up.weight | 0xd9b95620 | 0x7000000 |
133
+ | 34 | blk.11.ffn_norm.weight | 0xe0b95620 | 0x4000 |
134
+ | 35 | blk.11.attn_k.weight | 0xe0b99620 | 0x800000 |
135
+ | 36 | blk.11.attn_output.weight | 0xe1399620 | 0x2000000 |
136
+ | 37 | blk.11.attn_q.weight | 0xe3399620 | 0x2000000 |
137
+ | 38 | blk.11.attn_v.weight | 0xe5399620 | 0x800000 |
138
+ | 39 | blk.12.attn_norm.weight | 0xe5b99620 | 0x4000 |
139
+ | 40 | blk.12.ffn_down.weight | 0xe5b9d620 | 0x7000000 |
140
+ | 41 | blk.12.ffn_gate.weight | 0xecb9d620 | 0x7000000 |
141
+ | 42 | blk.12.ffn_up.weight | 0xf3b9d620 | 0x7000000 |
142
+ | 43 | blk.12.ffn_norm.weight | 0xfab9d620 | 0x4000 |
143
+ | 44 | blk.12.attn_k.weight | 0xfaba1620 | 0x800000 |
144
+ | 45 | blk.12.attn_output.weight | 0xfb3a1620 | 0x2000000 |
145
+ | 46 | blk.12.attn_q.weight | 0xfd3a1620 | 0x2000000 |
146
+ | 47 | blk.12.attn_v.weight | 0xff3a1620 | 0x800000 |
147
+ | 48 | blk.13.attn_norm.weight | 0xffba1620 | 0x4000 |
148
+ | 49 | blk.13.ffn_down.weight | 0xffba5620 | 0x7000000 |
149
+ | 50 | blk.13.ffn_gate.weight | 0x106ba5620 | 0x7000000 |
150
+ | 51 | blk.13.ffn_up.weight | 0x10dba5620 | 0x7000000 |
151
+ | 52 | blk.13.ffn_norm.weight | 0x114ba5620 | 0x4000 |
152
+ | 53 | blk.13.attn_k.weight | 0x114ba9620 | 0x800000 |
153
+ | 54 | blk.13.attn_output.weight | 0x1153a9620 | 0x2000000 |
154
+ | 55 | blk.13.attn_q.weight | 0x1173a9620 | 0x2000000 |
155
+ | 56 | blk.13.attn_v.weight | 0x1193a9620 | 0x800000 |
156
+ | 57 | blk.14.attn_norm.weight | 0x119ba9620 | 0x4000 |
157
+ | 58 | blk.14.ffn_down.weight | 0x119bad620 | 0x7000000 |
158
+ | 59 | blk.14.ffn_gate.weight | 0x120bad620 | 0x7000000 |
159
+ | 60 | blk.14.ffn_up.weight | 0x127bad620 | 0x7000000 |
160
+ | 61 | blk.14.ffn_norm.weight | 0x12ebad620 | 0x4000 |
161
+ | 62 | blk.14.attn_k.weight | 0x12ebb1620 | 0x800000 |
162
+ | 63 | blk.14.attn_output.weight | 0x12f3b1620 | 0x2000000 |
163
+ | 64 | blk.14.attn_q.weight | 0x1313b1620 | 0x2000000 |
164
+ | 65 | blk.14.attn_v.weight | 0x1333b1620 | 0x800000 |
165
+ | 66 | blk.15.attn_norm.weight | 0x133bb1620 | 0x4000 |
166
+ | 67 | blk.15.ffn_down.weight | 0x133bb5620 | 0x7000000 |
167
+ | 68 | blk.15.ffn_gate.weight | 0x13abb5620 | 0x7000000 |
168
+ | 69 | blk.15.ffn_up.weight | 0x141bb5620 | 0x7000000 |
169
+ | 70 | blk.15.ffn_norm.weight | 0x148bb5620 | 0x4000 |
170
+ | 71 | blk.15.attn_k.weight | 0x148bb9620 | 0x800000 |
171
+ | 72 | blk.15.attn_output.weight | 0x1493b9620 | 0x2000000 |
172
+ | 73 | blk.15.attn_q.weight | 0x14b3b9620 | 0x2000000 |
173
+ | 74 | blk.15.attn_v.weight | 0x14d3b9620 | 0x800000 |
174
+ | 75 | blk.16.attn_norm.weight | 0x14dbb9620 | 0x4000 |
175
+ | 76 | blk.16.ffn_down.weight | 0x14dbbd620 | 0x7000000 |
176
+ | 77 | blk.16.ffn_gate.weight | 0x154bbd620 | 0x7000000 |
177
+ | 78 | blk.16.ffn_up.weight | 0x15bbbd620 | 0x7000000 |
178
+ | 79 | blk.16.ffn_norm.weight | 0x162bbd620 | 0x4000 |
179
+ | 80 | blk.16.attn_k.weight | 0x162bc1620 | 0x800000 |
180
+ | 81 | blk.16.attn_output.weight | 0x1633c1620 | 0x2000000 |
181
+ | 82 | blk.16.attn_q.weight | 0x1653c1620 | 0x2000000 |
182
+ | 83 | blk.16.attn_v.weight | 0x1673c1620 | 0x800000 |
183
+ | 84 | blk.17.attn_norm.weight | 0x167bc1620 | 0x4000 |
184
+ | 85 | blk.17.ffn_down.weight | 0x167bc5620 | 0x7000000 |
185
+ | 86 | blk.17.ffn_gate.weight | 0x16ebc5620 | 0x7000000 |
186
+ | 87 | blk.17.ffn_up.weight | 0x175bc5620 | 0x7000000 |
187
+ | 88 | blk.17.ffn_norm.weight | 0x17cbc5620 | 0x4000 |
188
+ | 89 | blk.17.attn_k.weight | 0x17cbc9620 | 0x800000 |
189
+ | 90 | blk.17.attn_output.weight | 0x17d3c9620 | 0x2000000 |
190
+ | 91 | blk.17.attn_q.weight | 0x17f3c9620 | 0x2000000 |
191
+ | 92 | blk.17.attn_v.weight | 0x1813c9620 | 0x800000 |
192
+ | 93 | blk.18.attn_norm.weight | 0x181bc9620 | 0x4000 |
193
+ | 94 | blk.18.ffn_down.weight | 0x181bcd620 | 0x7000000 |
194
+ | 95 | blk.18.ffn_gate.weight | 0x188bcd620 | 0x7000000 |
195
+ | 96 | blk.18.ffn_up.weight | 0x18fbcd620 | 0x7000000 |
196
+ | 97 | blk.18.ffn_norm.weight | 0x196bcd620 | 0x4000 |
197
+ | 98 | blk.18.attn_k.weight | 0x196bd1620 | 0x800000 |
198
+ | 99 | blk.18.attn_output.weight | 0x1973d1620 | 0x2000000 |
199
+ | 100 | blk.18.attn_q.weight | 0x1993d1620 | 0x2000000 |
200
+ | 101 | blk.18.attn_v.weight | 0x19b3d1620 | 0x800000 |
201
+ | 102 | blk.19.attn_norm.weight | 0x19bbd1620 | 0x4000 |
202
+ | 103 | blk.19.ffn_down.weight | 0x19bbd5620 | 0x7000000 |
203
+ | 104 | blk.19.ffn_gate.weight | 0x1a2bd5620 | 0x7000000 |
204
+ | 105 | blk.19.ffn_up.weight | 0x1a9bd5620 | 0x7000000 |
205
+ | 106 | blk.19.ffn_norm.weight | 0x1b0bd5620 | 0x4000 |
206
+ | 107 | blk.19.attn_k.weight | 0x1b0bd9620 | 0x800000 |
207
+ | 108 | blk.19.attn_output.weight | 0x1b13d9620 | 0x2000000 |
208
+ | 109 | blk.19.attn_q.weight | 0x1b33d9620 | 0x2000000 |
209
+ | 110 | blk.19.attn_v.weight | 0x1b53d9620 | 0x800000 |
210
+ | 111 | blk.2.attn_norm.weight | 0x1b5bd9620 | 0x4000 |
211
+ | 112 | blk.2.ffn_down.weight | 0x1b5bdd620 | 0x7000000 |
212
+ | 113 | blk.2.ffn_gate.weight | 0x1bcbdd620 | 0x7000000 |
213
+ | 114 | blk.2.ffn_up.weight | 0x1c3bdd620 | 0x7000000 |
214
+ | 115 | blk.2.ffn_norm.weight | 0x1cabdd620 | 0x4000 |
215
+ | 116 | blk.2.attn_k.weight | 0x1cabe1620 | 0x800000 |
216
+ | 117 | blk.2.attn_output.weight | 0x1cb3e1620 | 0x2000000 |
217
+ | 118 | blk.2.attn_q.weight | 0x1cd3e1620 | 0x2000000 |
218
+ | 119 | blk.2.attn_v.weight | 0x1cf3e1620 | 0x800000 |
219
+ | 120 | blk.20.attn_norm.weight | 0x1cfbe1620 | 0x4000 |
220
+ | 121 | blk.20.ffn_down.weight | 0x1cfbe5620 | 0x7000000 |
221
+ | 122 | blk.20.ffn_gate.weight | 0x1d6be5620 | 0x7000000 |
222
+ | 123 | blk.20.ffn_up.weight | 0x1ddbe5620 | 0x7000000 |
223
+ | 124 | blk.20.ffn_norm.weight | 0x1e4be5620 | 0x4000 |
224
+ | 125 | blk.20.attn_k.weight | 0x1e4be9620 | 0x800000 |
225
+ | 126 | blk.20.attn_output.weight | 0x1e53e9620 | 0x2000000 |
226
+ | 127 | blk.20.attn_q.weight | 0x1e73e9620 | 0x2000000 |
227
+ | 128 | blk.20.attn_v.weight | 0x1e93e9620 | 0x800000 |
228
+ | 129 | blk.21.attn_norm.weight | 0x1e9be9620 | 0x4000 |
229
+ | 130 | blk.21.ffn_down.weight | 0x1e9bed620 | 0x7000000 |
230
+ | 131 | blk.21.ffn_gate.weight | 0x1f0bed620 | 0x7000000 |
231
+ | 132 | blk.21.ffn_up.weight | 0x1f7bed620 | 0x7000000 |
232
+ | 133 | blk.21.ffn_norm.weight | 0x1febed620 | 0x4000 |
233
+ | 134 | blk.21.attn_k.weight | 0x1febf1620 | 0x800000 |
234
+ | 135 | blk.21.attn_output.weight | 0x1ff3f1620 | 0x2000000 |
235
+ | 136 | blk.21.attn_q.weight | 0x2013f1620 | 0x2000000 |
236
+ | 137 | blk.21.attn_v.weight | 0x2033f1620 | 0x800000 |
237
+ | 138 | blk.22.attn_norm.weight | 0x203bf1620 | 0x4000 |
238
+ | 139 | blk.22.ffn_down.weight | 0x203bf5620 | 0x7000000 |
239
+ | 140 | blk.22.ffn_gate.weight | 0x20abf5620 | 0x7000000 |
240
+ | 141 | blk.22.ffn_up.weight | 0x211bf5620 | 0x7000000 |
241
+ | 142 | blk.22.ffn_norm.weight | 0x218bf5620 | 0x4000 |
242
+ | 143 | blk.22.attn_k.weight | 0x218bf9620 | 0x800000 |
243
+ | 144 | blk.22.attn_output.weight | 0x2193f9620 | 0x2000000 |
244
+ | 145 | blk.22.attn_q.weight | 0x21b3f9620 | 0x2000000 |
245
+ | 146 | blk.22.attn_v.weight | 0x21d3f9620 | 0x800000 |
246
+ | 147 | blk.23.attn_norm.weight | 0x21dbf9620 | 0x4000 |
247
+ | 148 | blk.23.ffn_down.weight | 0x21dbfd620 | 0x7000000 |
248
+ | 149 | blk.23.ffn_gate.weight | 0x224bfd620 | 0x7000000 |
249
+ | 150 | blk.23.ffn_up.weight | 0x22bbfd620 | 0x7000000 |
250
+ | 151 | blk.23.ffn_norm.weight | 0x232bfd620 | 0x4000 |
251
+ | 152 | blk.23.attn_k.weight | 0x232c01620 | 0x800000 |
252
+ | 153 | blk.23.attn_output.weight | 0x233401620 | 0x2000000 |
253
+ | 154 | blk.23.attn_q.weight | 0x235401620 | 0x2000000 |
254
+ | 155 | blk.23.attn_v.weight | 0x237401620 | 0x800000 |
255
+ | 156 | blk.24.attn_norm.weight | 0x237c01620 | 0x4000 |
256
+ | 157 | blk.24.ffn_down.weight | 0x237c05620 | 0x7000000 |
257
+ | 158 | blk.24.ffn_gate.weight | 0x23ec05620 | 0x7000000 |
258
+ | 159 | blk.24.ffn_up.weight | 0x245c05620 | 0x7000000 |
259
+ | 160 | blk.24.ffn_norm.weight | 0x24cc05620 | 0x4000 |
260
+ | 161 | blk.24.attn_k.weight | 0x24cc09620 | 0x800000 |
261
+ | 162 | blk.24.attn_output.weight | 0x24d409620 | 0x2000000 |
262
+ | 163 | blk.24.attn_q.weight | 0x24f409620 | 0x2000000 |
263
+ | 164 | blk.24.attn_v.weight | 0x251409620 | 0x800000 |
264
+ | 165 | blk.25.attn_norm.weight | 0x251c09620 | 0x4000 |
265
+ | 166 | blk.25.ffn_down.weight | 0x251c0d620 | 0x7000000 |
266
+ | 167 | blk.25.ffn_gate.weight | 0x258c0d620 | 0x7000000 |
267
+ | 168 | blk.25.ffn_up.weight | 0x25fc0d620 | 0x7000000 |
268
+ | 169 | blk.25.ffn_norm.weight | 0x266c0d620 | 0x4000 |
269
+ | 170 | blk.25.attn_k.weight | 0x266c11620 | 0x800000 |
270
+ | 171 | blk.25.attn_output.weight | 0x267411620 | 0x2000000 |
271
+ | 172 | blk.25.attn_q.weight | 0x269411620 | 0x2000000 |
272
+ | 173 | blk.25.attn_v.weight | 0x26b411620 | 0x800000 |
273
+ | 174 | blk.26.attn_norm.weight | 0x26bc11620 | 0x4000 |
274
+ | 175 | blk.26.ffn_down.weight | 0x26bc15620 | 0x7000000 |
275
+ | 176 | blk.26.ffn_gate.weight | 0x272c15620 | 0x7000000 |
276
+ | 177 | blk.26.ffn_up.weight | 0x279c15620 | 0x7000000 |
277
+ | 178 | blk.26.ffn_norm.weight | 0x280c15620 | 0x4000 |
278
+ | 179 | blk.26.attn_k.weight | 0x280c19620 | 0x800000 |
279
+ | 180 | blk.26.attn_output.weight | 0x281419620 | 0x2000000 |
280
+ | 181 | blk.26.attn_q.weight | 0x283419620 | 0x2000000 |
281
+ | 182 | blk.26.attn_v.weight | 0x285419620 | 0x800000 |
282
+ | 183 | blk.27.attn_norm.weight | 0x285c19620 | 0x4000 |
283
+ | 184 | blk.27.ffn_down.weight | 0x285c1d620 | 0x7000000 |
284
+ | 185 | blk.27.ffn_gate.weight | 0x28cc1d620 | 0x7000000 |
285
+ | 186 | blk.27.ffn_up.weight | 0x293c1d620 | 0x7000000 |
286
+ | 187 | blk.27.ffn_norm.weight | 0x29ac1d620 | 0x4000 |
287
+ | 188 | blk.27.attn_k.weight | 0x29ac21620 | 0x800000 |
288
+ | 189 | blk.27.attn_output.weight | 0x29b421620 | 0x2000000 |
289
+ | 190 | blk.27.attn_q.weight | 0x29d421620 | 0x2000000 |
290
+ | 191 | blk.27.attn_v.weight | 0x29f421620 | 0x800000 |
291
+ | 192 | blk.28.attn_norm.weight | 0x29fc21620 | 0x4000 |
292
+ | 193 | blk.28.ffn_down.weight | 0x29fc25620 | 0x7000000 |
293
+ | 194 | blk.28.ffn_gate.weight | 0x2a6c25620 | 0x7000000 |
294
+ | 195 | blk.28.ffn_up.weight | 0x2adc25620 | 0x7000000 |
295
+ | 196 | blk.28.ffn_norm.weight | 0x2b4c25620 | 0x4000 |
296
+ | 197 | blk.28.attn_k.weight | 0x2b4c29620 | 0x800000 |
297
+ | 198 | blk.28.attn_output.weight | 0x2b5429620 | 0x2000000 |
298
+ | 199 | blk.28.attn_q.weight | 0x2b7429620 | 0x2000000 |
299
+ | 200 | blk.28.attn_v.weight | 0x2b9429620 | 0x800000 |
300
+ | 201 | blk.29.attn_norm.weight | 0x2b9c29620 | 0x4000 |
301
+ | 202 | blk.29.ffn_down.weight | 0x2b9c2d620 | 0x7000000 |
302
+ | 203 | blk.29.ffn_gate.weight | 0x2c0c2d620 | 0x7000000 |
303
+ | 204 | blk.29.ffn_up.weight | 0x2c7c2d620 | 0x7000000 |
304
+ | 205 | blk.29.ffn_norm.weight | 0x2cec2d620 | 0x4000 |
305
+ | 206 | blk.29.attn_k.weight | 0x2cec31620 | 0x800000 |
306
+ | 207 | blk.29.attn_output.weight | 0x2cf431620 | 0x2000000 |
307
+ | 208 | blk.29.attn_q.weight | 0x2d1431620 | 0x2000000 |
308
+ | 209 | blk.29.attn_v.weight | 0x2d3431620 | 0x800000 |
309
+ | 210 | blk.3.attn_norm.weight | 0x2d3c31620 | 0x4000 |
310
+ | 211 | blk.3.ffn_down.weight | 0x2d3c35620 | 0x7000000 |
311
+ | 212 | blk.3.ffn_gate.weight | 0x2dac35620 | 0x7000000 |
312
+ | 213 | blk.3.ffn_up.weight | 0x2e1c35620 | 0x7000000 |
313
+ | 214 | blk.3.ffn_norm.weight | 0x2e8c35620 | 0x4000 |
314
+ | 215 | blk.3.attn_k.weight | 0x2e8c39620 | 0x800000 |
315
+ | 216 | blk.3.attn_output.weight | 0x2e9439620 | 0x2000000 |
316
+ | 217 | blk.3.attn_q.weight | 0x2eb439620 | 0x2000000 |
317
+ | 218 | blk.3.attn_v.weight | 0x2ed439620 | 0x800000 |
318
+ | 219 | blk.30.attn_norm.weight | 0x2edc39620 | 0x4000 |
319
+ | 220 | blk.30.ffn_down.weight | 0x2edc3d620 | 0x7000000 |
320
+ | 221 | blk.30.ffn_gate.weight | 0x2f4c3d620 | 0x7000000 |
321
+ | 222 | blk.30.ffn_up.weight | 0x2fbc3d620 | 0x7000000 |
322
+ | 223 | blk.30.ffn_norm.weight | 0x302c3d620 | 0x4000 |
323
+ | 224 | blk.30.attn_k.weight | 0x302c41620 | 0x800000 |
324
+ | 225 | blk.30.attn_output.weight | 0x303441620 | 0x2000000 |
325
+ | 226 | blk.30.attn_q.weight | 0x305441620 | 0x2000000 |
326
+ | 227 | blk.30.attn_v.weight | 0x307441620 | 0x800000 |
327
+ | 228 | blk.31.attn_norm.weight | 0x307c41620 | 0x4000 |
328
+ | 229 | blk.31.ffn_down.weight | 0x307c45620 | 0x7000000 |
329
+ | 230 | blk.31.ffn_gate.weight | 0x30ec45620 | 0x7000000 |
330
+ | 231 | blk.31.ffn_up.weight | 0x315c45620 | 0x7000000 |
331
+ | 232 | blk.31.ffn_norm.weight | 0x31cc45620 | 0x4000 |
332
+ | 233 | blk.31.attn_k.weight | 0x31cc49620 | 0x800000 |
333
+ | 234 | blk.31.attn_output.weight | 0x31d449620 | 0x2000000 |
334
+ | 235 | blk.31.attn_q.weight | 0x31f449620 | 0x2000000 |
335
+ | 236 | blk.31.attn_v.weight | 0x321449620 | 0x800000 |
336
+ | 237 | blk.4.attn_norm.weight | 0x321c49620 | 0x4000 |
337
+ | 238 | blk.4.ffn_down.weight | 0x321c4d620 | 0x7000000 |
338
+ | 239 | blk.4.ffn_gate.weight | 0x328c4d620 | 0x7000000 |
339
+ | 240 | blk.4.ffn_up.weight | 0x32fc4d620 | 0x7000000 |
340
+ | 241 | blk.4.ffn_norm.weight | 0x336c4d620 | 0x4000 |
341
+ | 242 | blk.4.attn_k.weight | 0x336c51620 | 0x800000 |
342
+ | 243 | blk.4.attn_output.weight | 0x337451620 | 0x2000000 |
343
+ | 244 | blk.4.attn_q.weight | 0x339451620 | 0x2000000 |
344
+ | 245 | blk.4.attn_v.weight | 0x33b451620 | 0x800000 |
345
+ | 246 | blk.5.attn_norm.weight | 0x33bc51620 | 0x4000 |
346
+ | 247 | blk.5.ffn_down.weight | 0x33bc55620 | 0x7000000 |
347
+ | 248 | blk.5.ffn_gate.weight | 0x342c55620 | 0x7000000 |
348
+ | 249 | blk.5.ffn_up.weight | 0x349c55620 | 0x7000000 |
349
+ | 250 | blk.5.ffn_norm.weight | 0x350c55620 | 0x4000 |
350
+ | 251 | blk.5.attn_k.weight | 0x350c59620 | 0x800000 |
351
+ | 252 | blk.5.attn_output.weight | 0x351459620 | 0x2000000 |
352
+ | 253 | blk.5.attn_q.weight | 0x353459620 | 0x2000000 |
353
+ | 254 | blk.5.attn_v.weight | 0x355459620 | 0x800000 |
354
+ | 255 | blk.6.attn_norm.weight | 0x355c59620 | 0x4000 |
355
+ | 256 | blk.6.ffn_down.weight | 0x355c5d620 | 0x7000000 |
356
+ | 257 | blk.6.ffn_gate.weight | 0x35cc5d620 | 0x7000000 |
357
+ | 258 | blk.6.ffn_up.weight | 0x363c5d620 | 0x7000000 |
358
+ | 259 | blk.6.ffn_norm.weight | 0x36ac5d620 | 0x4000 |
359
+ | 260 | blk.6.attn_k.weight | 0x36ac61620 | 0x800000 |
360
+ | 261 | blk.6.attn_output.weight | 0x36b461620 | 0x2000000 |
361
+ | 262 | blk.6.attn_q.weight | 0x36d461620 | 0x2000000 |
362
+ | 263 | blk.6.attn_v.weight | 0x36f461620 | 0x800000 |
363
+ | 264 | blk.7.attn_norm.weight | 0x36fc61620 | 0x4000 |
364
+ | 265 | blk.7.ffn_down.weight | 0x36fc65620 | 0x7000000 |
365
+ | 266 | blk.7.ffn_gate.weight | 0x376c65620 | 0x7000000 |
366
+ | 267 | blk.7.ffn_up.weight | 0x37dc65620 | 0x7000000 |
367
+ | 268 | blk.7.ffn_norm.weight | 0x384c65620 | 0x4000 |
368
+ | 269 | blk.7.attn_k.weight | 0x384c69620 | 0x800000 |
369
+ | 270 | blk.7.attn_output.weight | 0x385469620 | 0x2000000 |
370
+ | 271 | blk.7.attn_q.weight | 0x387469620 | 0x2000000 |
371
+ | 272 | blk.7.attn_v.weight | 0x389469620 | 0x800000 |
372
+ | 273 | blk.8.attn_norm.weight | 0x389c69620 | 0x4000 |
373
+ | 274 | blk.8.ffn_down.weight | 0x389c6d620 | 0x7000000 |
374
+ | 275 | blk.8.ffn_gate.weight | 0x390c6d620 | 0x7000000 |
375
+ | 276 | blk.8.ffn_up.weight | 0x397c6d620 | 0x7000000 |
376
+ | 277 | blk.8.ffn_norm.weight | 0x39ec6d620 | 0x4000 |
377
+ | 278 | blk.8.attn_k.weight | 0x39ec71620 | 0x800000 |
378
+ | 279 | blk.8.attn_output.weight | 0x39f471620 | 0x2000000 |
379
+ | 280 | blk.8.attn_q.weight | 0x3a1471620 | 0x2000000 |
380
+ | 281 | blk.8.attn_v.weight | 0x3a3471620 | 0x800000 |
381
+ | 282 | blk.9.attn_norm.weight | 0x3a3c71620 | 0x4000 |
382
+ | 283 | blk.9.ffn_down.weight | 0x3a3c75620 | 0x7000000 |
383
+ | 284 | blk.9.ffn_gate.weight | 0x3aac75620 | 0x7000000 |
384
+ | 285 | blk.9.ffn_up.weight | 0x3b1c75620 | 0x7000000 |
385
+ | 286 | blk.9.ffn_norm.weight | 0x3b8c75620 | 0x4000 |
386
+ | 287 | blk.9.attn_k.weight | 0x3b8c79620 | 0x800000 |
387
+ | 288 | blk.9.attn_output.weight | 0x3b9479620 | 0x2000000 |
388
+ | 289 | blk.9.attn_q.weight | 0x3bb479620 | 0x2000000 |
389
+ | 290 | blk.9.attn_v.weight | 0x3bd479620 | 0x800000 |
390
+ | 291 | output_norm.weight | 0x3bdc79620 | 0x4000 |
391
+
392
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
393
+
394
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
395
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----|
396
+ | 0 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
397
+ | 1 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | F16 |
398
+ | 2 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | F16 |
399
+ | 291 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
400
+
401
+ - Total elements in base: ( ~1B) 1050677312
402
+ - Percentage of total elements: 13.08%
403
+
404
+
405
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
406
+
407
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
408
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
409
+ | 3 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
410
+ | 4 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
411
+ | 5 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
412
+ | 6 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
413
+ | 7 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
414
+ | 8 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
415
+ | 9 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
416
+ | 10 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
417
+ | 11 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
418
+
419
+ - Total elements in blk.0: (~218M) 218112000
420
+ - Percentage of total elements: 2.72%
421
+
422
+
423
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
424
+
425
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
426
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
427
+ | 12 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
428
+ | 13 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
429
+ | 14 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
430
+ | 15 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
431
+ | 16 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
432
+ | 17 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
433
+ | 18 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
434
+ | 19 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
435
+ | 20 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
436
+
437
+ - Total elements in blk.1: (~218M) 218112000
438
+ - Percentage of total elements: 2.72%
439
+
440
+
441
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
442
+
443
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
444
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
445
+ | 21 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
446
+ | 22 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
447
+ | 23 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
448
+ | 24 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
449
+ | 25 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
450
+ | 26 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
451
+ | 27 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
452
+ | 28 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
453
+ | 29 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
454
+
455
+ - Total elements in blk.10: (~218M) 218112000
456
+ - Percentage of total elements: 2.72%
457
+
458
+
459
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
460
+
461
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
462
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
463
+ | 30 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
464
+ | 31 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
465
+ | 32 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
466
+ | 33 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
467
+ | 34 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
468
+ | 35 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
469
+ | 36 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
470
+ | 37 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
471
+ | 38 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
472
+
473
+ - Total elements in blk.11: (~218M) 218112000
474
+ - Percentage of total elements: 2.72%
475
+
476
+
477
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
478
+
479
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
480
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
481
+ | 39 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
482
+ | 40 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
483
+ | 41 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
484
+ | 42 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
485
+ | 43 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
486
+ | 44 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
487
+ | 45 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
488
+ | 46 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
489
+ | 47 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
490
+
491
+ - Total elements in blk.12: (~218M) 218112000
492
+ - Percentage of total elements: 2.72%
493
+
494
+
495
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
496
+
497
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
498
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
499
+ | 48 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
500
+ | 49 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
501
+ | 50 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
502
+ | 51 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
503
+ | 52 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
504
+ | 53 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
505
+ | 54 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
506
+ | 55 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
507
+ | 56 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
508
+
509
+ - Total elements in blk.13: (~218M) 218112000
510
+ - Percentage of total elements: 2.72%
511
+
512
+
513
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
514
+
515
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
516
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
517
+ | 57 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
518
+ | 58 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
519
+ | 59 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
520
+ | 60 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
521
+ | 61 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
522
+ | 62 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
523
+ | 63 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
524
+ | 64 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
525
+ | 65 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
526
+
527
+ - Total elements in blk.14: (~218M) 218112000
528
+ - Percentage of total elements: 2.72%
529
+
530
+
531
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
532
+
533
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
534
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
535
+ | 66 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
536
+ | 67 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
537
+ | 68 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
538
+ | 69 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
539
+ | 70 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
540
+ | 71 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
541
+ | 72 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
542
+ | 73 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
543
+ | 74 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
544
+
545
+ - Total elements in blk.15: (~218M) 218112000
546
+ - Percentage of total elements: 2.72%
547
+
548
+
549
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
550
+
551
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
552
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
553
+ | 75 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
554
+ | 76 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
555
+ | 77 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
556
+ | 78 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
557
+ | 79 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
558
+ | 80 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
559
+ | 81 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
560
+ | 82 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
561
+ | 83 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
562
+
563
+ - Total elements in blk.16: (~218M) 218112000
564
+ - Percentage of total elements: 2.72%
565
+
566
+
567
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
568
+
569
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
570
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
571
+ | 84 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
572
+ | 85 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
573
+ | 86 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
574
+ | 87 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
575
+ | 88 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
576
+ | 89 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
577
+ | 90 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
578
+ | 91 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
579
+ | 92 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
580
+
581
+ - Total elements in blk.17: (~218M) 218112000
582
+ - Percentage of total elements: 2.72%
583
+
584
+
585
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
586
+
587
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
588
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
589
+ | 93 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
590
+ | 94 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
591
+ | 95 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
592
+ | 96 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
593
+ | 97 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
594
+ | 98 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
595
+ | 99 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
596
+ | 100 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
597
+ | 101 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
598
+
599
+ - Total elements in blk.18: (~218M) 218112000
600
+ - Percentage of total elements: 2.72%
601
+
602
+
603
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
604
+
605
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
606
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
607
+ | 102 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
608
+ | 103 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
609
+ | 104 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
610
+ | 105 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
611
+ | 106 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
612
+ | 107 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
613
+ | 108 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
614
+ | 109 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
615
+ | 110 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
616
+
617
+ - Total elements in blk.19: (~218M) 218112000
618
+ - Percentage of total elements: 2.72%
619
+
620
+
621
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
622
+
623
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
624
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
625
+ | 111 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
626
+ | 112 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
627
+ | 113 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
628
+ | 114 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
629
+ | 115 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
630
+ | 116 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
631
+ | 117 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
632
+ | 118 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
633
+ | 119 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
634
+
635
+ - Total elements in blk.2: (~218M) 218112000
636
+ - Percentage of total elements: 2.72%
637
+
638
+
639
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
640
+
641
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
642
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
643
+ | 120 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
644
+ | 121 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
645
+ | 122 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
646
+ | 123 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
647
+ | 124 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
648
+ | 125 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
649
+ | 126 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
650
+ | 127 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
651
+ | 128 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
652
+
653
+ - Total elements in blk.20: (~218M) 218112000
654
+ - Percentage of total elements: 2.72%
655
+
656
+
657
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
658
+
659
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
660
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
661
+ | 129 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
662
+ | 130 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
663
+ | 131 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
664
+ | 132 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
665
+ | 133 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
666
+ | 134 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
667
+ | 135 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
668
+ | 136 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
669
+ | 137 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
670
+
671
+ - Total elements in blk.21: (~218M) 218112000
672
+ - Percentage of total elements: 2.72%
673
+
674
+
675
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
676
+
677
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
678
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
679
+ | 138 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
680
+ | 139 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
681
+ | 140 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
682
+ | 141 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
683
+ | 142 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
684
+ | 143 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
685
+ | 144 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
686
+ | 145 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
687
+ | 146 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
688
+
689
+ - Total elements in blk.22: (~218M) 218112000
690
+ - Percentage of total elements: 2.72%
691
+
692
+
693
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
694
+
695
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
696
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
697
+ | 147 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
698
+ | 148 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
699
+ | 149 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
700
+ | 150 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
701
+ | 151 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
702
+ | 152 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
703
+ | 153 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
704
+ | 154 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
705
+ | 155 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
706
+
707
+ - Total elements in blk.23: (~218M) 218112000
708
+ - Percentage of total elements: 2.72%
709
+
710
+
711
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
712
+
713
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
714
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
715
+ | 156 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
716
+ | 157 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
717
+ | 158 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
718
+ | 159 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
719
+ | 160 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
720
+ | 161 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
721
+ | 162 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
722
+ | 163 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
723
+ | 164 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
724
+
725
+ - Total elements in blk.24: (~218M) 218112000
726
+ - Percentage of total elements: 2.72%
727
+
728
+
729
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
730
+
731
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
732
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
733
+ | 165 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
734
+ | 166 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
735
+ | 167 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
736
+ | 168 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
737
+ | 169 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
738
+ | 170 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
739
+ | 171 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
740
+ | 172 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
741
+ | 173 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
742
+
743
+ - Total elements in blk.25: (~218M) 218112000
744
+ - Percentage of total elements: 2.72%
745
+
746
+
747
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
748
+
749
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
750
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
751
+ | 174 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
752
+ | 175 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
753
+ | 176 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
754
+ | 177 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
755
+ | 178 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
756
+ | 179 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
757
+ | 180 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
758
+ | 181 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
759
+ | 182 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
760
+
761
+ - Total elements in blk.26: (~218M) 218112000
762
+ - Percentage of total elements: 2.72%
763
+
764
+
765
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
766
+
767
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
768
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
769
+ | 183 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
770
+ | 184 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
771
+ | 185 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
772
+ | 186 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
773
+ | 187 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
774
+ | 188 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
775
+ | 189 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
776
+ | 190 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
777
+ | 191 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
778
+
779
+ - Total elements in blk.27: (~218M) 218112000
780
+ - Percentage of total elements: 2.72%
781
+
782
+
783
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
784
+
785
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
786
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
787
+ | 192 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
788
+ | 193 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
789
+ | 194 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
790
+ | 195 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
791
+ | 196 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
792
+ | 197 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
793
+ | 198 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
794
+ | 199 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
795
+ | 200 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
796
+
797
+ - Total elements in blk.28: (~218M) 218112000
798
+ - Percentage of total elements: 2.72%
799
+
800
+
801
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
802
+
803
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
804
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
805
+ | 201 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
806
+ | 202 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
807
+ | 203 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
808
+ | 204 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
809
+ | 205 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
810
+ | 206 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
811
+ | 207 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
812
+ | 208 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
813
+ | 209 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
814
+
815
+ - Total elements in blk.29: (~218M) 218112000
816
+ - Percentage of total elements: 2.72%
817
+
818
+
819
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
820
+
821
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
822
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
823
+ | 210 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
824
+ | 211 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
825
+ | 212 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
826
+ | 213 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
827
+ | 214 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
828
+ | 215 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
829
+ | 216 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
830
+ | 217 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
831
+ | 218 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
832
+
833
+ - Total elements in blk.3: (~218M) 218112000
834
+ - Percentage of total elements: 2.72%
835
+
836
+
837
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
838
+
839
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
840
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
841
+ | 219 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
842
+ | 220 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
843
+ | 221 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
844
+ | 222 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
845
+ | 223 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
846
+ | 224 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
847
+ | 225 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
848
+ | 226 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
849
+ | 227 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
850
+
851
+ - Total elements in blk.30: (~218M) 218112000
852
+ - Percentage of total elements: 2.72%
853
+
854
+
855
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
856
+
857
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
858
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
859
+ | 228 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
860
+ | 229 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
861
+ | 230 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
862
+ | 231 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
863
+ | 232 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
864
+ | 233 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
865
+ | 234 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
866
+ | 235 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
867
+ | 236 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
868
+
869
+ - Total elements in blk.31: (~218M) 218112000
870
+ - Percentage of total elements: 2.72%
871
+
872
+
873
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
874
+
875
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
876
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
877
+ | 237 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
878
+ | 238 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
879
+ | 239 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
880
+ | 240 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
881
+ | 241 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
882
+ | 242 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
883
+ | 243 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
884
+ | 244 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
885
+ | 245 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
886
+
887
+ - Total elements in blk.4: (~218M) 218112000
888
+ - Percentage of total elements: 2.72%
889
+
890
+
891
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
892
+
893
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
894
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
895
+ | 246 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
896
+ | 247 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
897
+ | 248 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
898
+ | 249 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
899
+ | 250 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
900
+ | 251 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
901
+ | 252 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
902
+ | 253 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
903
+ | 254 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
904
+
905
+ - Total elements in blk.5: (~218M) 218112000
906
+ - Percentage of total elements: 2.72%
907
+
908
+
909
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
910
+
911
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
912
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
913
+ | 255 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
914
+ | 256 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
915
+ | 257 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
916
+ | 258 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
917
+ | 259 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
918
+ | 260 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
919
+ | 261 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
920
+ | 262 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
921
+ | 263 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
922
+
923
+ - Total elements in blk.6: (~218M) 218112000
924
+ - Percentage of total elements: 2.72%
925
+
926
+
927
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
928
+
929
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
930
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
931
+ | 264 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
932
+ | 265 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
933
+ | 266 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
934
+ | 267 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
935
+ | 268 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
936
+ | 269 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
937
+ | 270 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
938
+ | 271 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
939
+ | 272 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
940
+
941
+ - Total elements in blk.7: (~218M) 218112000
942
+ - Percentage of total elements: 2.72%
943
+
944
+
945
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
946
+
947
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
948
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
949
+ | 273 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
950
+ | 274 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
951
+ | 275 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
952
+ | 276 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
953
+ | 277 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
954
+ | 278 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
955
+ | 279 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
956
+ | 280 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
957
+ | 281 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
958
+
959
+ - Total elements in blk.8: (~218M) 218112000
960
+ - Percentage of total elements: 2.72%
961
+
962
+
963
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
964
+
965
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
966
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
967
+ | 282 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
968
+ | 283 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | F16 |
969
+ | 284 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
970
+ | 285 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | F16 |
971
+ | 286 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
972
+ | 287 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
973
+ | 288 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
974
+ | 289 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | F16 |
975
+ | 290 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
976
+
977
+ - Total elements in blk.9: (~218M) 218112000
978
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-IQ3_M.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-IQ3_M.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 27 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-IQ3\_M.gguf - GGUF Internal File Dump](#watt-tool-8b-iq3_mgguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0xd746000 |
104
+ | 1 | output_norm.weight | 0xdebf620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0xdec3620 | 0x100 |
106
+ | 3 | token_embd.weight | 0xdec3720 | 0xd746000 |
107
+ | 4 | blk.0.attn_k.weight | 0x1b609720 | 0x188000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x1b791720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x1b795720 | 0x900000 |
110
+ | 7 | blk.0.attn_q.weight | 0x1c095720 | 0x620000 |
111
+ | 8 | blk.0.attn_v.weight | 0x1c6b5720 | 0x1b8000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x1c86d720 | 0x1f80000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x1e7ed720 | 0x1570000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x1fd5d720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x1fd61720 | 0x1570000 |
116
+ | 13 | blk.1.attn_k.weight | 0x212d1720 | 0x188000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x21459720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x2145d720 | 0x900000 |
119
+ | 16 | blk.1.attn_q.weight | 0x21d5d720 | 0x620000 |
120
+ | 17 | blk.1.attn_v.weight | 0x2237d720 | 0x1b8000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x22535720 | 0x1f80000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x244b5720 | 0x1570000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x25a25720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x25a29720 | 0x1570000 |
125
+ | 22 | blk.2.attn_k.weight | 0x26f99720 | 0x188000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x27121720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x27125720 | 0x900000 |
128
+ | 25 | blk.2.attn_q.weight | 0x27a25720 | 0x620000 |
129
+ | 26 | blk.2.attn_v.weight | 0x28045720 | 0x1b8000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x281fd720 | 0x1f80000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x2a17d720 | 0x1570000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x2b6ed720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x2b6f1720 | 0x1570000 |
134
+ | 31 | blk.3.attn_k.weight | 0x2cc61720 | 0x188000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x2cde9720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x2cded720 | 0x900000 |
137
+ | 34 | blk.3.attn_q.weight | 0x2d6ed720 | 0x620000 |
138
+ | 35 | blk.3.attn_v.weight | 0x2dd0d720 | 0x1b8000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x2dec5720 | 0x1f80000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x2fe45720 | 0x1570000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x313b5720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x313b9720 | 0x1570000 |
143
+ | 40 | blk.4.attn_k.weight | 0x32929720 | 0x188000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x32ab1720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x32ab5720 | 0x900000 |
146
+ | 43 | blk.4.attn_q.weight | 0x333b5720 | 0x620000 |
147
+ | 44 | blk.4.attn_v.weight | 0x339d5720 | 0x1b8000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x33b8d720 | 0x1f80000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x35b0d720 | 0x1570000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x3707d720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x37081720 | 0x1570000 |
152
+ | 49 | blk.5.attn_k.weight | 0x385f1720 | 0x188000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x38779720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x3877d720 | 0x900000 |
155
+ | 52 | blk.5.attn_q.weight | 0x3907d720 | 0x620000 |
156
+ | 53 | blk.5.attn_v.weight | 0x3969d720 | 0x1b8000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x39855720 | 0x1f80000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x3b7d5720 | 0x1570000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x3cd45720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x3cd49720 | 0x1570000 |
161
+ | 58 | blk.6.attn_k.weight | 0x3e2b9720 | 0x188000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x3e441720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x3e445720 | 0x900000 |
164
+ | 61 | blk.6.attn_q.weight | 0x3ed45720 | 0x620000 |
165
+ | 62 | blk.6.attn_v.weight | 0x3f365720 | 0x1b8000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x3f51d720 | 0x1f80000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x4149d720 | 0x1570000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x42a0d720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x42a11720 | 0x1570000 |
170
+ | 67 | blk.7.attn_k.weight | 0x43f81720 | 0x188000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x44109720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x4410d720 | 0x900000 |
173
+ | 70 | blk.7.attn_q.weight | 0x44a0d720 | 0x620000 |
174
+ | 71 | blk.7.attn_v.weight | 0x4502d720 | 0x1b8000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x451e5720 | 0x1f80000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x47165720 | 0x1570000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x486d5720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x486d9720 | 0x1570000 |
179
+ | 76 | blk.8.attn_k.weight | 0x49c49720 | 0x188000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x49dd1720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x49dd5720 | 0x900000 |
182
+ | 79 | blk.8.attn_q.weight | 0x4a6d5720 | 0x620000 |
183
+ | 80 | blk.8.attn_v.weight | 0x4acf5720 | 0x1b8000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x4aead720 | 0x1f80000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x4ce2d720 | 0x1570000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x4e39d720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x4e3a1720 | 0x1570000 |
188
+ | 85 | blk.9.attn_k.weight | 0x4f911720 | 0x188000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x4fa99720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x4fa9d720 | 0x900000 |
191
+ | 88 | blk.9.attn_q.weight | 0x5039d720 | 0x620000 |
192
+ | 89 | blk.9.attn_v.weight | 0x509bd720 | 0x1b8000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x50b75720 | 0x1f80000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x52af5720 | 0x1570000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0x54065720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0x54069720 | 0x1570000 |
197
+ | 94 | blk.10.attn_k.weight | 0x555d9720 | 0x188000 |
198
+ | 95 | blk.10.attn_norm.weight | 0x55761720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0x55765720 | 0x900000 |
200
+ | 97 | blk.10.attn_q.weight | 0x56065720 | 0x620000 |
201
+ | 98 | blk.10.attn_v.weight | 0x56685720 | 0x1b8000 |
202
+ | 99 | blk.10.ffn_down.weight | 0x5683d720 | 0x1f80000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0x587bd720 | 0x1570000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0x59d2d720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0x59d31720 | 0x1570000 |
206
+ | 103 | blk.11.attn_k.weight | 0x5b2a1720 | 0x188000 |
207
+ | 104 | blk.11.attn_norm.weight | 0x5b429720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0x5b42d720 | 0x900000 |
209
+ | 106 | blk.11.attn_q.weight | 0x5bd2d720 | 0x620000 |
210
+ | 107 | blk.11.attn_v.weight | 0x5c34d720 | 0x1b8000 |
211
+ | 108 | blk.11.ffn_down.weight | 0x5c505720 | 0x1f80000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0x5e485720 | 0x1570000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0x5f9f5720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0x5f9f9720 | 0x1570000 |
215
+ | 112 | blk.12.attn_k.weight | 0x60f69720 | 0x188000 |
216
+ | 113 | blk.12.attn_norm.weight | 0x610f1720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0x610f5720 | 0x900000 |
218
+ | 115 | blk.12.attn_q.weight | 0x619f5720 | 0x620000 |
219
+ | 116 | blk.12.attn_v.weight | 0x62015720 | 0x1b8000 |
220
+ | 117 | blk.12.ffn_down.weight | 0x621cd720 | 0x1f80000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0x6414d720 | 0x1570000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0x656bd720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0x656c1720 | 0x1570000 |
224
+ | 121 | blk.13.attn_k.weight | 0x66c31720 | 0x1b8000 |
225
+ | 122 | blk.13.attn_norm.weight | 0x66de9720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0x66ded720 | 0x900000 |
227
+ | 124 | blk.13.attn_q.weight | 0x676ed720 | 0x6e0000 |
228
+ | 125 | blk.13.attn_v.weight | 0x67dcd720 | 0x240000 |
229
+ | 126 | blk.13.ffn_down.weight | 0x6800d720 | 0x1f80000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0x69f8d720 | 0x1570000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0x6b4fd720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0x6b501720 | 0x1570000 |
233
+ | 130 | blk.14.attn_k.weight | 0x6ca71720 | 0x1b8000 |
234
+ | 131 | blk.14.attn_norm.weight | 0x6cc29720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0x6cc2d720 | 0x900000 |
236
+ | 133 | blk.14.attn_q.weight | 0x6d52d720 | 0x6e0000 |
237
+ | 134 | blk.14.attn_v.weight | 0x6dc0d720 | 0x240000 |
238
+ | 135 | blk.14.ffn_down.weight | 0x6de4d720 | 0x1f80000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0x6fdcd720 | 0x1570000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0x7133d720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0x71341720 | 0x1570000 |
242
+ | 139 | blk.15.attn_k.weight | 0x728b1720 | 0x188000 |
243
+ | 140 | blk.15.attn_norm.weight | 0x72a39720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0x72a3d720 | 0x900000 |
245
+ | 142 | blk.15.attn_q.weight | 0x7333d720 | 0x620000 |
246
+ | 143 | blk.15.attn_v.weight | 0x7395d720 | 0x1b8000 |
247
+ | 144 | blk.15.ffn_down.weight | 0x73b15720 | 0x1f80000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0x75a95720 | 0x1570000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0x77005720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0x77009720 | 0x1570000 |
251
+ | 148 | blk.16.attn_k.weight | 0x78579720 | 0x1b8000 |
252
+ | 149 | blk.16.attn_norm.weight | 0x78731720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0x78735720 | 0x900000 |
254
+ | 151 | blk.16.attn_q.weight | 0x79035720 | 0x6e0000 |
255
+ | 152 | blk.16.attn_v.weight | 0x79715720 | 0x240000 |
256
+ | 153 | blk.16.ffn_down.weight | 0x79955720 | 0x1f80000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0x7b8d5720 | 0x1810000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0x7d0e5720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0x7d0e9720 | 0x1810000 |
260
+ | 157 | blk.17.attn_k.weight | 0x7e8f9720 | 0x188000 |
261
+ | 158 | blk.17.attn_norm.weight | 0x7ea81720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0x7ea85720 | 0x900000 |
263
+ | 160 | blk.17.attn_q.weight | 0x7f385720 | 0x620000 |
264
+ | 161 | blk.17.attn_v.weight | 0x7f9a5720 | 0x1b8000 |
265
+ | 162 | blk.17.ffn_down.weight | 0x7fb5d720 | 0x1f80000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0x81add720 | 0x1810000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0x832ed720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0x832f1720 | 0x1810000 |
269
+ | 166 | blk.18.attn_k.weight | 0x84b01720 | 0x1b8000 |
270
+ | 167 | blk.18.attn_norm.weight | 0x84cb9720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0x84cbd720 | 0x900000 |
272
+ | 169 | blk.18.attn_q.weight | 0x855bd720 | 0x6e0000 |
273
+ | 170 | blk.18.attn_v.weight | 0x85c9d720 | 0x240000 |
274
+ | 171 | blk.18.ffn_down.weight | 0x85edd720 | 0x1f80000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0x87e5d720 | 0x1810000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0x8966d720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0x89671720 | 0x1810000 |
278
+ | 175 | blk.19.attn_k.weight | 0x8ae81720 | 0x1b8000 |
279
+ | 176 | blk.19.attn_norm.weight | 0x8b039720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0x8b03d720 | 0x900000 |
281
+ | 178 | blk.19.attn_q.weight | 0x8b93d720 | 0x6e0000 |
282
+ | 179 | blk.19.attn_v.weight | 0x8c01d720 | 0x240000 |
283
+ | 180 | blk.19.ffn_down.weight | 0x8c25d720 | 0x1f80000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0x8e1dd720 | 0x1810000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0x8f9ed720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0x8f9f1720 | 0x1810000 |
287
+ | 184 | blk.20.attn_k.weight | 0x91201720 | 0x1b8000 |
288
+ | 185 | blk.20.attn_norm.weight | 0x913b9720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0x913bd720 | 0x900000 |
290
+ | 187 | blk.20.attn_q.weight | 0x91cbd720 | 0x6e0000 |
291
+ | 188 | blk.20.attn_v.weight | 0x9239d720 | 0x240000 |
292
+ | 189 | blk.20.ffn_down.weight | 0x925dd720 | 0x1f80000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0x9455d720 | 0x1810000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0x95d6d720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0x95d71720 | 0x1810000 |
296
+ | 193 | blk.21.attn_k.weight | 0x97581720 | 0x1b8000 |
297
+ | 194 | blk.21.attn_norm.weight | 0x97739720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0x9773d720 | 0x900000 |
299
+ | 196 | blk.21.attn_q.weight | 0x9803d720 | 0x6e0000 |
300
+ | 197 | blk.21.attn_v.weight | 0x9871d720 | 0x240000 |
301
+ | 198 | blk.21.ffn_down.weight | 0x9895d720 | 0x1f80000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0x9a8dd720 | 0x1810000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0x9c0ed720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0x9c0f1720 | 0x1810000 |
305
+ | 202 | blk.22.attn_k.weight | 0x9d901720 | 0x1b8000 |
306
+ | 203 | blk.22.attn_norm.weight | 0x9dab9720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0x9dabd720 | 0x900000 |
308
+ | 205 | blk.22.attn_q.weight | 0x9e3bd720 | 0x6e0000 |
309
+ | 206 | blk.22.attn_v.weight | 0x9ea9d720 | 0x240000 |
310
+ | 207 | blk.22.ffn_down.weight | 0x9ecdd720 | 0x1f80000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0xa0c5d720 | 0x1810000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0xa246d720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0xa2471720 | 0x1810000 |
314
+ | 211 | blk.23.attn_k.weight | 0xa3c81720 | 0x1b8000 |
315
+ | 212 | blk.23.attn_norm.weight | 0xa3e39720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0xa3e3d720 | 0x900000 |
317
+ | 214 | blk.23.attn_q.weight | 0xa473d720 | 0x6e0000 |
318
+ | 215 | blk.23.attn_v.weight | 0xa4e1d720 | 0x240000 |
319
+ | 216 | blk.23.ffn_down.weight | 0xa505d720 | 0x1f80000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0xa6fdd720 | 0x1810000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0xa87ed720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0xa87f1720 | 0x1810000 |
323
+ | 220 | blk.24.attn_k.weight | 0xaa001720 | 0x1b8000 |
324
+ | 221 | blk.24.attn_norm.weight | 0xaa1b9720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0xaa1bd720 | 0x900000 |
326
+ | 223 | blk.24.attn_q.weight | 0xaaabd720 | 0x6e0000 |
327
+ | 224 | blk.24.attn_v.weight | 0xab19d720 | 0x240000 |
328
+ | 225 | blk.24.ffn_down.weight | 0xab3dd720 | 0x1f80000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0xad35d720 | 0x1810000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0xaeb6d720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0xaeb71720 | 0x1810000 |
332
+ | 229 | blk.25.attn_k.weight | 0xb0381720 | 0x1b8000 |
333
+ | 230 | blk.25.attn_norm.weight | 0xb0539720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0xb053d720 | 0x900000 |
335
+ | 232 | blk.25.attn_q.weight | 0xb0e3d720 | 0x6e0000 |
336
+ | 233 | blk.25.attn_v.weight | 0xb151d720 | 0x240000 |
337
+ | 234 | blk.25.ffn_down.weight | 0xb175d720 | 0x1f80000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0xb36dd720 | 0x1810000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0xb4eed720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0xb4ef1720 | 0x1810000 |
341
+ | 238 | blk.26.attn_k.weight | 0xb6701720 | 0x1b8000 |
342
+ | 239 | blk.26.attn_norm.weight | 0xb68b9720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0xb68bd720 | 0x900000 |
344
+ | 241 | blk.26.attn_q.weight | 0xb71bd720 | 0x6e0000 |
345
+ | 242 | blk.26.attn_v.weight | 0xb789d720 | 0x240000 |
346
+ | 243 | blk.26.ffn_down.weight | 0xb7add720 | 0x1f80000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0xb9a5d720 | 0x1810000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0xbb26d720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0xbb271720 | 0x1810000 |
350
+ | 247 | blk.27.attn_k.weight | 0xbca81720 | 0x1b8000 |
351
+ | 248 | blk.27.attn_norm.weight | 0xbcc39720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0xbcc3d720 | 0x900000 |
353
+ | 250 | blk.27.attn_q.weight | 0xbd53d720 | 0x6e0000 |
354
+ | 251 | blk.27.attn_v.weight | 0xbdc1d720 | 0x240000 |
355
+ | 252 | blk.27.ffn_down.weight | 0xbde5d720 | 0x1f80000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0xbfddd720 | 0x1810000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0xc15ed720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0xc15f1720 | 0x1810000 |
359
+ | 256 | blk.28.attn_k.weight | 0xc2e01720 | 0x1b8000 |
360
+ | 257 | blk.28.attn_norm.weight | 0xc2fb9720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0xc2fbd720 | 0x900000 |
362
+ | 259 | blk.28.attn_q.weight | 0xc38bd720 | 0x6e0000 |
363
+ | 260 | blk.28.attn_v.weight | 0xc3f9d720 | 0x240000 |
364
+ | 261 | blk.28.ffn_down.weight | 0xc41dd720 | 0x1f80000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0xc615d720 | 0x1810000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0xc796d720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0xc7971720 | 0x1810000 |
368
+ | 265 | blk.29.attn_k.weight | 0xc9181720 | 0x1b8000 |
369
+ | 266 | blk.29.attn_norm.weight | 0xc9339720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0xc933d720 | 0x900000 |
371
+ | 268 | blk.29.attn_q.weight | 0xc9c3d720 | 0x6e0000 |
372
+ | 269 | blk.29.attn_v.weight | 0xca31d720 | 0x240000 |
373
+ | 270 | blk.29.ffn_down.weight | 0xca55d720 | 0x1f80000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0xcc4dd720 | 0x1810000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0xcdced720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0xcdcf1720 | 0x1810000 |
377
+ | 274 | blk.30.attn_k.weight | 0xcf501720 | 0x1b8000 |
378
+ | 275 | blk.30.attn_norm.weight | 0xcf6b9720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0xcf6bd720 | 0x900000 |
380
+ | 277 | blk.30.attn_q.weight | 0xcffbd720 | 0x6e0000 |
381
+ | 278 | blk.30.attn_v.weight | 0xd069d720 | 0x240000 |
382
+ | 279 | blk.30.ffn_down.weight | 0xd08dd720 | 0x1f80000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0xd285d720 | 0x1810000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0xd406d720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0xd4071720 | 0x1810000 |
386
+ | 283 | blk.31.attn_k.weight | 0xd5881720 | 0x188000 |
387
+ | 284 | blk.31.attn_norm.weight | 0xd5a09720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0xd5a0d720 | 0x900000 |
389
+ | 286 | blk.31.attn_q.weight | 0xd630d720 | 0x620000 |
390
+ | 287 | blk.31.attn_v.weight | 0xd692d720 | 0x1b8000 |
391
+ | 288 | blk.31.ffn_down.weight | 0xd6ae5720 | 0x1f80000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0xd8a65720 | 0x1810000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0xda275720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0xda279720 | 0x1810000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:------|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | IQ3_S |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | IQ3_S |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-IQ3_S.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-IQ3_S.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 26 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-IQ3\_S.gguf - GGUF Internal File Dump](#watt-tool-8b-iq3_sgguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0xbfca000 |
104
+ | 1 | output_norm.weight | 0xc743620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0xc747620 | 0x100 |
106
+ | 3 | token_embd.weight | 0xc747720 | 0xbfca000 |
107
+ | 4 | blk.0.attn_k.weight | 0x18711720 | 0x188000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x18899720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x1889d720 | 0x6e0000 |
110
+ | 7 | blk.0.attn_q.weight | 0x18f7d720 | 0x620000 |
111
+ | 8 | blk.0.attn_v.weight | 0x1959d720 | 0x1b8000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x19755720 | 0x1810000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x1af65720 | 0x1570000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x1c4d5720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x1c4d9720 | 0x1570000 |
116
+ | 13 | blk.1.attn_k.weight | 0x1da49720 | 0x188000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x1dbd1720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x1dbd5720 | 0x6e0000 |
119
+ | 16 | blk.1.attn_q.weight | 0x1e2b5720 | 0x620000 |
120
+ | 17 | blk.1.attn_v.weight | 0x1e8d5720 | 0x1b8000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x1ea8d720 | 0x1f80000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x20a0d720 | 0x1570000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x21f7d720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x21f81720 | 0x1570000 |
125
+ | 22 | blk.2.attn_k.weight | 0x234f1720 | 0x188000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x23679720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x2367d720 | 0x6e0000 |
128
+ | 25 | blk.2.attn_q.weight | 0x23d5d720 | 0x620000 |
129
+ | 26 | blk.2.attn_v.weight | 0x2437d720 | 0x1b8000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x24535720 | 0x1810000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x25d45720 | 0x1570000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x272b5720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x272b9720 | 0x1570000 |
134
+ | 31 | blk.3.attn_k.weight | 0x28829720 | 0x188000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x289b1720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x289b5720 | 0x6e0000 |
137
+ | 34 | blk.3.attn_q.weight | 0x29095720 | 0x620000 |
138
+ | 35 | blk.3.attn_v.weight | 0x296b5720 | 0x1b8000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x2986d720 | 0x1810000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x2b07d720 | 0x1570000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x2c5ed720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x2c5f1720 | 0x1570000 |
143
+ | 40 | blk.4.attn_k.weight | 0x2db61720 | 0x188000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x2dce9720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x2dced720 | 0x6e0000 |
146
+ | 43 | blk.4.attn_q.weight | 0x2e3cd720 | 0x620000 |
147
+ | 44 | blk.4.attn_v.weight | 0x2e9ed720 | 0x1b8000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x2eba5720 | 0x1810000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x303b5720 | 0x1570000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x31925720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x31929720 | 0x1570000 |
152
+ | 49 | blk.5.attn_k.weight | 0x32e99720 | 0x188000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x33021720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x33025720 | 0x6e0000 |
155
+ | 52 | blk.5.attn_q.weight | 0x33705720 | 0x620000 |
156
+ | 53 | blk.5.attn_v.weight | 0x33d25720 | 0x1b8000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x33edd720 | 0x1810000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x356ed720 | 0x1570000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x36c5d720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x36c61720 | 0x1570000 |
161
+ | 58 | blk.6.attn_k.weight | 0x381d1720 | 0x188000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x38359720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x3835d720 | 0x6e0000 |
164
+ | 61 | blk.6.attn_q.weight | 0x38a3d720 | 0x620000 |
165
+ | 62 | blk.6.attn_v.weight | 0x3905d720 | 0x1b8000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x39215720 | 0x1810000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x3aa25720 | 0x1570000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x3bf95720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x3bf99720 | 0x1570000 |
170
+ | 67 | blk.7.attn_k.weight | 0x3d509720 | 0x188000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x3d691720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x3d695720 | 0x6e0000 |
173
+ | 70 | blk.7.attn_q.weight | 0x3dd75720 | 0x620000 |
174
+ | 71 | blk.7.attn_v.weight | 0x3e395720 | 0x1b8000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x3e54d720 | 0x1810000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x3fd5d720 | 0x1570000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x412cd720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x412d1720 | 0x1570000 |
179
+ | 76 | blk.8.attn_k.weight | 0x42841720 | 0x188000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x429c9720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x429cd720 | 0x6e0000 |
182
+ | 79 | blk.8.attn_q.weight | 0x430ad720 | 0x620000 |
183
+ | 80 | blk.8.attn_v.weight | 0x436cd720 | 0x1b8000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x43885720 | 0x1810000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x45095720 | 0x1570000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x46605720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x46609720 | 0x1570000 |
188
+ | 85 | blk.9.attn_k.weight | 0x47b79720 | 0x188000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x47d01720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x47d05720 | 0x6e0000 |
191
+ | 88 | blk.9.attn_q.weight | 0x483e5720 | 0x620000 |
192
+ | 89 | blk.9.attn_v.weight | 0x48a05720 | 0x1b8000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x48bbd720 | 0x1810000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x4a3cd720 | 0x1570000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0x4b93d720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0x4b941720 | 0x1570000 |
197
+ | 94 | blk.10.attn_k.weight | 0x4ceb1720 | 0x188000 |
198
+ | 95 | blk.10.attn_norm.weight | 0x4d039720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0x4d03d720 | 0x6e0000 |
200
+ | 97 | blk.10.attn_q.weight | 0x4d71d720 | 0x620000 |
201
+ | 98 | blk.10.attn_v.weight | 0x4dd3d720 | 0x1b8000 |
202
+ | 99 | blk.10.ffn_down.weight | 0x4def5720 | 0x1810000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0x4f705720 | 0x1570000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0x50c75720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0x50c79720 | 0x1570000 |
206
+ | 103 | blk.11.attn_k.weight | 0x521e9720 | 0x188000 |
207
+ | 104 | blk.11.attn_norm.weight | 0x52371720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0x52375720 | 0x6e0000 |
209
+ | 106 | blk.11.attn_q.weight | 0x52a55720 | 0x620000 |
210
+ | 107 | blk.11.attn_v.weight | 0x53075720 | 0x1b8000 |
211
+ | 108 | blk.11.ffn_down.weight | 0x5322d720 | 0x1810000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0x54a3d720 | 0x1570000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0x55fad720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0x55fb1720 | 0x1570000 |
215
+ | 112 | blk.12.attn_k.weight | 0x57521720 | 0x188000 |
216
+ | 113 | blk.12.attn_norm.weight | 0x576a9720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0x576ad720 | 0x6e0000 |
218
+ | 115 | blk.12.attn_q.weight | 0x57d8d720 | 0x620000 |
219
+ | 116 | blk.12.attn_v.weight | 0x583ad720 | 0x1b8000 |
220
+ | 117 | blk.12.ffn_down.weight | 0x58565720 | 0x1810000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0x59d75720 | 0x1570000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0x5b2e5720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0x5b2e9720 | 0x1570000 |
224
+ | 121 | blk.13.attn_k.weight | 0x5c859720 | 0x1b8000 |
225
+ | 122 | blk.13.attn_norm.weight | 0x5ca11720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0x5ca15720 | 0x6e0000 |
227
+ | 124 | blk.13.attn_q.weight | 0x5d0f5720 | 0x6e0000 |
228
+ | 125 | blk.13.attn_v.weight | 0x5d7d5720 | 0x1b8000 |
229
+ | 126 | blk.13.ffn_down.weight | 0x5d98d720 | 0x1810000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0x5f19d720 | 0x1570000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0x6070d720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0x60711720 | 0x1570000 |
233
+ | 130 | blk.14.attn_k.weight | 0x61c81720 | 0x1b8000 |
234
+ | 131 | blk.14.attn_norm.weight | 0x61e39720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0x61e3d720 | 0x6e0000 |
236
+ | 133 | blk.14.attn_q.weight | 0x6251d720 | 0x6e0000 |
237
+ | 134 | blk.14.attn_v.weight | 0x62bfd720 | 0x1b8000 |
238
+ | 135 | blk.14.ffn_down.weight | 0x62db5720 | 0x1810000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0x645c5720 | 0x1570000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0x65b35720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0x65b39720 | 0x1570000 |
242
+ | 139 | blk.15.attn_k.weight | 0x670a9720 | 0x188000 |
243
+ | 140 | blk.15.attn_norm.weight | 0x67231720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0x67235720 | 0x6e0000 |
245
+ | 142 | blk.15.attn_q.weight | 0x67915720 | 0x620000 |
246
+ | 143 | blk.15.attn_v.weight | 0x67f35720 | 0x1b8000 |
247
+ | 144 | blk.15.ffn_down.weight | 0x680ed720 | 0x1810000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0x698fd720 | 0x1570000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0x6ae6d720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0x6ae71720 | 0x1570000 |
251
+ | 148 | blk.16.attn_k.weight | 0x6c3e1720 | 0x1b8000 |
252
+ | 149 | blk.16.attn_norm.weight | 0x6c599720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0x6c59d720 | 0x6e0000 |
254
+ | 151 | blk.16.attn_q.weight | 0x6cc7d720 | 0x6e0000 |
255
+ | 152 | blk.16.attn_v.weight | 0x6d35d720 | 0x1b8000 |
256
+ | 153 | blk.16.ffn_down.weight | 0x6d515720 | 0x1810000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0x6ed25720 | 0x1810000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0x70535720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0x70539720 | 0x1810000 |
260
+ | 157 | blk.17.attn_k.weight | 0x71d49720 | 0x188000 |
261
+ | 158 | blk.17.attn_norm.weight | 0x71ed1720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0x71ed5720 | 0x6e0000 |
263
+ | 160 | blk.17.attn_q.weight | 0x725b5720 | 0x620000 |
264
+ | 161 | blk.17.attn_v.weight | 0x72bd5720 | 0x1b8000 |
265
+ | 162 | blk.17.ffn_down.weight | 0x72d8d720 | 0x1f80000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0x74d0d720 | 0x1810000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0x7651d720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0x76521720 | 0x1810000 |
269
+ | 166 | blk.18.attn_k.weight | 0x77d31720 | 0x1b8000 |
270
+ | 167 | blk.18.attn_norm.weight | 0x77ee9720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0x77eed720 | 0x6e0000 |
272
+ | 169 | blk.18.attn_q.weight | 0x785cd720 | 0x6e0000 |
273
+ | 170 | blk.18.attn_v.weight | 0x78cad720 | 0x1b8000 |
274
+ | 171 | blk.18.ffn_down.weight | 0x78e65720 | 0x1f80000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0x7ade5720 | 0x1810000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0x7c5f5720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0x7c5f9720 | 0x1810000 |
278
+ | 175 | blk.19.attn_k.weight | 0x7de09720 | 0x1b8000 |
279
+ | 176 | blk.19.attn_norm.weight | 0x7dfc1720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0x7dfc5720 | 0x6e0000 |
281
+ | 178 | blk.19.attn_q.weight | 0x7e6a5720 | 0x6e0000 |
282
+ | 179 | blk.19.attn_v.weight | 0x7ed85720 | 0x1b8000 |
283
+ | 180 | blk.19.ffn_down.weight | 0x7ef3d720 | 0x1f80000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0x80ebd720 | 0x1810000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0x826cd720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0x826d1720 | 0x1810000 |
287
+ | 184 | blk.20.attn_k.weight | 0x83ee1720 | 0x1b8000 |
288
+ | 185 | blk.20.attn_norm.weight | 0x84099720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0x8409d720 | 0x6e0000 |
290
+ | 187 | blk.20.attn_q.weight | 0x8477d720 | 0x6e0000 |
291
+ | 188 | blk.20.attn_v.weight | 0x84e5d720 | 0x1b8000 |
292
+ | 189 | blk.20.ffn_down.weight | 0x85015720 | 0x1f80000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0x86f95720 | 0x1810000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0x887a5720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0x887a9720 | 0x1810000 |
296
+ | 193 | blk.21.attn_k.weight | 0x89fb9720 | 0x1b8000 |
297
+ | 194 | blk.21.attn_norm.weight | 0x8a171720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0x8a175720 | 0x6e0000 |
299
+ | 196 | blk.21.attn_q.weight | 0x8a855720 | 0x6e0000 |
300
+ | 197 | blk.21.attn_v.weight | 0x8af35720 | 0x1b8000 |
301
+ | 198 | blk.21.ffn_down.weight | 0x8b0ed720 | 0x1f80000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0x8d06d720 | 0x1810000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0x8e87d720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0x8e881720 | 0x1810000 |
305
+ | 202 | blk.22.attn_k.weight | 0x90091720 | 0x1b8000 |
306
+ | 203 | blk.22.attn_norm.weight | 0x90249720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0x9024d720 | 0x6e0000 |
308
+ | 205 | blk.22.attn_q.weight | 0x9092d720 | 0x6e0000 |
309
+ | 206 | blk.22.attn_v.weight | 0x9100d720 | 0x1b8000 |
310
+ | 207 | blk.22.ffn_down.weight | 0x911c5720 | 0x1f80000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0x93145720 | 0x1810000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0x94955720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0x94959720 | 0x1810000 |
314
+ | 211 | blk.23.attn_k.weight | 0x96169720 | 0x1b8000 |
315
+ | 212 | blk.23.attn_norm.weight | 0x96321720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0x96325720 | 0x6e0000 |
317
+ | 214 | blk.23.attn_q.weight | 0x96a05720 | 0x6e0000 |
318
+ | 215 | blk.23.attn_v.weight | 0x970e5720 | 0x1b8000 |
319
+ | 216 | blk.23.ffn_down.weight | 0x9729d720 | 0x1f80000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0x9921d720 | 0x1810000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0x9aa2d720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0x9aa31720 | 0x1810000 |
323
+ | 220 | blk.24.attn_k.weight | 0x9c241720 | 0x1b8000 |
324
+ | 221 | blk.24.attn_norm.weight | 0x9c3f9720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0x9c3fd720 | 0x6e0000 |
326
+ | 223 | blk.24.attn_q.weight | 0x9cadd720 | 0x6e0000 |
327
+ | 224 | blk.24.attn_v.weight | 0x9d1bd720 | 0x1b8000 |
328
+ | 225 | blk.24.ffn_down.weight | 0x9d375720 | 0x1f80000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0x9f2f5720 | 0x1810000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0xa0b05720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0xa0b09720 | 0x1810000 |
332
+ | 229 | blk.25.attn_k.weight | 0xa2319720 | 0x1b8000 |
333
+ | 230 | blk.25.attn_norm.weight | 0xa24d1720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0xa24d5720 | 0x6e0000 |
335
+ | 232 | blk.25.attn_q.weight | 0xa2bb5720 | 0x6e0000 |
336
+ | 233 | blk.25.attn_v.weight | 0xa3295720 | 0x1b8000 |
337
+ | 234 | blk.25.ffn_down.weight | 0xa344d720 | 0x1f80000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0xa53cd720 | 0x1810000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0xa6bdd720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0xa6be1720 | 0x1810000 |
341
+ | 238 | blk.26.attn_k.weight | 0xa83f1720 | 0x1b8000 |
342
+ | 239 | blk.26.attn_norm.weight | 0xa85a9720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0xa85ad720 | 0x6e0000 |
344
+ | 241 | blk.26.attn_q.weight | 0xa8c8d720 | 0x6e0000 |
345
+ | 242 | blk.26.attn_v.weight | 0xa936d720 | 0x1b8000 |
346
+ | 243 | blk.26.ffn_down.weight | 0xa9525720 | 0x1f80000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0xab4a5720 | 0x1810000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0xaccb5720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0xaccb9720 | 0x1810000 |
350
+ | 247 | blk.27.attn_k.weight | 0xae4c9720 | 0x1b8000 |
351
+ | 248 | blk.27.attn_norm.weight | 0xae681720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0xae685720 | 0x6e0000 |
353
+ | 250 | blk.27.attn_q.weight | 0xaed65720 | 0x6e0000 |
354
+ | 251 | blk.27.attn_v.weight | 0xaf445720 | 0x1b8000 |
355
+ | 252 | blk.27.ffn_down.weight | 0xaf5fd720 | 0x1f80000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0xb157d720 | 0x1810000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0xb2d8d720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0xb2d91720 | 0x1810000 |
359
+ | 256 | blk.28.attn_k.weight | 0xb45a1720 | 0x1b8000 |
360
+ | 257 | blk.28.attn_norm.weight | 0xb4759720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0xb475d720 | 0x6e0000 |
362
+ | 259 | blk.28.attn_q.weight | 0xb4e3d720 | 0x6e0000 |
363
+ | 260 | blk.28.attn_v.weight | 0xb551d720 | 0x1b8000 |
364
+ | 261 | blk.28.ffn_down.weight | 0xb56d5720 | 0x1f80000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0xb7655720 | 0x1810000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0xb8e65720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0xb8e69720 | 0x1810000 |
368
+ | 265 | blk.29.attn_k.weight | 0xba679720 | 0x1b8000 |
369
+ | 266 | blk.29.attn_norm.weight | 0xba831720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0xba835720 | 0x6e0000 |
371
+ | 268 | blk.29.attn_q.weight | 0xbaf15720 | 0x6e0000 |
372
+ | 269 | blk.29.attn_v.weight | 0xbb5f5720 | 0x1b8000 |
373
+ | 270 | blk.29.ffn_down.weight | 0xbb7ad720 | 0x1f80000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0xbd72d720 | 0x1810000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0xbef3d720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0xbef41720 | 0x1810000 |
377
+ | 274 | blk.30.attn_k.weight | 0xc0751720 | 0x1b8000 |
378
+ | 275 | blk.30.attn_norm.weight | 0xc0909720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0xc090d720 | 0x6e0000 |
380
+ | 277 | blk.30.attn_q.weight | 0xc0fed720 | 0x6e0000 |
381
+ | 278 | blk.30.attn_v.weight | 0xc16cd720 | 0x1b8000 |
382
+ | 279 | blk.30.ffn_down.weight | 0xc1885720 | 0x1f80000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0xc3805720 | 0x1810000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0xc5015720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0xc5019720 | 0x1810000 |
386
+ | 283 | blk.31.attn_k.weight | 0xc6829720 | 0x188000 |
387
+ | 284 | blk.31.attn_norm.weight | 0xc69b1720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0xc69b5720 | 0x6e0000 |
389
+ | 286 | blk.31.attn_q.weight | 0xc7095720 | 0x620000 |
390
+ | 287 | blk.31.attn_v.weight | 0xc76b5720 | 0x1b8000 |
391
+ | 288 | blk.31.ffn_down.weight | 0xc786d720 | 0x1f80000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0xc97ed720 | 0x1810000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0xcaffd720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0xcb001720 | 0x1810000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:--------|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | IQ3_XXS |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | IQ3_XXS |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:--------|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_XXS |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:------|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ3_S |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:--------|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_XXS |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_XXS |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | IQ4_NL |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-IQ4_NL.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-IQ4_NL.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 25 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-IQ4\_NL.gguf - GGUF Internal File Dump](#watt-tool-8b-iq4_nlgguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0x119d0000 |
104
+ | 1 | output_norm.weight | 0x12149620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0x1214d620 | 0x100 |
106
+ | 3 | token_embd.weight | 0x1214d720 | 0xd746000 |
107
+ | 4 | blk.0.attn_k.weight | 0x1f893720 | 0x1b8000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x1fa4b720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x1fa4f720 | 0x900000 |
110
+ | 7 | blk.0.attn_q.weight | 0x2034f720 | 0x6e0000 |
111
+ | 8 | blk.0.attn_v.weight | 0x20a2f720 | 0x220000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x20c4f720 | 0x2680000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x232cf720 | 0x1810000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x24adf720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x24ae3720 | 0x1810000 |
116
+ | 13 | blk.1.attn_k.weight | 0x262f3720 | 0x1b8000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x264ab720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x264af720 | 0x900000 |
119
+ | 16 | blk.1.attn_q.weight | 0x26daf720 | 0x6e0000 |
120
+ | 17 | blk.1.attn_v.weight | 0x2748f720 | 0x220000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x276af720 | 0x2680000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x29d2f720 | 0x1810000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x2b53f720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x2b543720 | 0x1810000 |
125
+ | 22 | blk.2.attn_k.weight | 0x2cd53720 | 0x1b8000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x2cf0b720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x2cf0f720 | 0x900000 |
128
+ | 25 | blk.2.attn_q.weight | 0x2d80f720 | 0x6e0000 |
129
+ | 26 | blk.2.attn_v.weight | 0x2deef720 | 0x220000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x2e10f720 | 0x2680000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x3078f720 | 0x1810000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x31f9f720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x31fa3720 | 0x1810000 |
134
+ | 31 | blk.3.attn_k.weight | 0x337b3720 | 0x1b8000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x3396b720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x3396f720 | 0x900000 |
137
+ | 34 | blk.3.attn_q.weight | 0x3426f720 | 0x6e0000 |
138
+ | 35 | blk.3.attn_v.weight | 0x3494f720 | 0x220000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x34b6f720 | 0x2680000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x371ef720 | 0x1810000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x389ff720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x38a03720 | 0x1810000 |
143
+ | 40 | blk.4.attn_k.weight | 0x3a213720 | 0x1b8000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x3a3cb720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x3a3cf720 | 0x900000 |
146
+ | 43 | blk.4.attn_q.weight | 0x3accf720 | 0x6e0000 |
147
+ | 44 | blk.4.attn_v.weight | 0x3b3af720 | 0x220000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x3b5cf720 | 0x2680000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x3dc4f720 | 0x1810000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x3f45f720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x3f463720 | 0x1810000 |
152
+ | 49 | blk.5.attn_k.weight | 0x40c73720 | 0x1b8000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x40e2b720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x40e2f720 | 0x900000 |
155
+ | 52 | blk.5.attn_q.weight | 0x4172f720 | 0x6e0000 |
156
+ | 53 | blk.5.attn_v.weight | 0x41e0f720 | 0x220000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x4202f720 | 0x2680000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x446af720 | 0x1810000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x45ebf720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x45ec3720 | 0x1810000 |
161
+ | 58 | blk.6.attn_k.weight | 0x476d3720 | 0x1b8000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x4788b720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x4788f720 | 0x900000 |
164
+ | 61 | blk.6.attn_q.weight | 0x4818f720 | 0x6e0000 |
165
+ | 62 | blk.6.attn_v.weight | 0x4886f720 | 0x220000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x48a8f720 | 0x2680000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x4b10f720 | 0x1810000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x4c91f720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x4c923720 | 0x1810000 |
170
+ | 67 | blk.7.attn_k.weight | 0x4e133720 | 0x1b8000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x4e2eb720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x4e2ef720 | 0x900000 |
173
+ | 70 | blk.7.attn_q.weight | 0x4ebef720 | 0x6e0000 |
174
+ | 71 | blk.7.attn_v.weight | 0x4f2cf720 | 0x220000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x4f4ef720 | 0x2680000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x51b6f720 | 0x1810000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x5337f720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x53383720 | 0x1810000 |
179
+ | 76 | blk.8.attn_k.weight | 0x54b93720 | 0x1b8000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x54d4b720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x54d4f720 | 0x900000 |
182
+ | 79 | blk.8.attn_q.weight | 0x5564f720 | 0x6e0000 |
183
+ | 80 | blk.8.attn_v.weight | 0x55d2f720 | 0x220000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x55f4f720 | 0x2680000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x585cf720 | 0x1810000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x59ddf720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x59de3720 | 0x1810000 |
188
+ | 85 | blk.9.attn_k.weight | 0x5b5f3720 | 0x1b8000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x5b7ab720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x5b7af720 | 0x900000 |
191
+ | 88 | blk.9.attn_q.weight | 0x5c0af720 | 0x6e0000 |
192
+ | 89 | blk.9.attn_v.weight | 0x5c78f720 | 0x220000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x5c9af720 | 0x2680000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x5f02f720 | 0x1810000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0x6083f720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0x60843720 | 0x1810000 |
197
+ | 94 | blk.10.attn_k.weight | 0x62053720 | 0x1b8000 |
198
+ | 95 | blk.10.attn_norm.weight | 0x6220b720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0x6220f720 | 0x900000 |
200
+ | 97 | blk.10.attn_q.weight | 0x62b0f720 | 0x6e0000 |
201
+ | 98 | blk.10.attn_v.weight | 0x631ef720 | 0x220000 |
202
+ | 99 | blk.10.ffn_down.weight | 0x6340f720 | 0x2680000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0x65a8f720 | 0x1810000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0x6729f720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0x672a3720 | 0x1810000 |
206
+ | 103 | blk.11.attn_k.weight | 0x68ab3720 | 0x1b8000 |
207
+ | 104 | blk.11.attn_norm.weight | 0x68c6b720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0x68c6f720 | 0x900000 |
209
+ | 106 | blk.11.attn_q.weight | 0x6956f720 | 0x6e0000 |
210
+ | 107 | blk.11.attn_v.weight | 0x69c4f720 | 0x220000 |
211
+ | 108 | blk.11.ffn_down.weight | 0x69e6f720 | 0x2680000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0x6c4ef720 | 0x1810000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0x6dcff720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0x6dd03720 | 0x1810000 |
215
+ | 112 | blk.12.attn_k.weight | 0x6f513720 | 0x1b8000 |
216
+ | 113 | blk.12.attn_norm.weight | 0x6f6cb720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0x6f6cf720 | 0x900000 |
218
+ | 115 | blk.12.attn_q.weight | 0x6ffcf720 | 0x6e0000 |
219
+ | 116 | blk.12.attn_v.weight | 0x706af720 | 0x220000 |
220
+ | 117 | blk.12.ffn_down.weight | 0x708cf720 | 0x2680000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0x72f4f720 | 0x1810000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0x7475f720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0x74763720 | 0x1810000 |
224
+ | 121 | blk.13.attn_k.weight | 0x75f73720 | 0x240000 |
225
+ | 122 | blk.13.attn_norm.weight | 0x761b3720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0x761b7720 | 0x900000 |
227
+ | 124 | blk.13.attn_q.weight | 0x76ab7720 | 0x900000 |
228
+ | 125 | blk.13.attn_v.weight | 0x773b7720 | 0x240000 |
229
+ | 126 | blk.13.ffn_down.weight | 0x775f7720 | 0x2680000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0x79c77720 | 0x1810000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0x7b487720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0x7b48b720 | 0x1810000 |
233
+ | 130 | blk.14.attn_k.weight | 0x7cc9b720 | 0x240000 |
234
+ | 131 | blk.14.attn_norm.weight | 0x7cedb720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0x7cedf720 | 0x900000 |
236
+ | 133 | blk.14.attn_q.weight | 0x7d7df720 | 0x900000 |
237
+ | 134 | blk.14.attn_v.weight | 0x7e0df720 | 0x240000 |
238
+ | 135 | blk.14.ffn_down.weight | 0x7e31f720 | 0x2680000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0x8099f720 | 0x1810000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0x821af720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0x821b3720 | 0x1810000 |
242
+ | 139 | blk.15.attn_k.weight | 0x839c3720 | 0x1b8000 |
243
+ | 140 | blk.15.attn_norm.weight | 0x83b7b720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0x83b7f720 | 0x900000 |
245
+ | 142 | blk.15.attn_q.weight | 0x8447f720 | 0x6e0000 |
246
+ | 143 | blk.15.attn_v.weight | 0x84b5f720 | 0x220000 |
247
+ | 144 | blk.15.ffn_down.weight | 0x84d7f720 | 0x2680000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0x873ff720 | 0x1810000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0x88c0f720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0x88c13720 | 0x1810000 |
251
+ | 148 | blk.16.attn_k.weight | 0x8a423720 | 0x240000 |
252
+ | 149 | blk.16.attn_norm.weight | 0x8a663720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0x8a667720 | 0x900000 |
254
+ | 151 | blk.16.attn_q.weight | 0x8af67720 | 0x900000 |
255
+ | 152 | blk.16.attn_v.weight | 0x8b867720 | 0x240000 |
256
+ | 153 | blk.16.ffn_down.weight | 0x8baa7720 | 0x2680000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0x8e127720 | 0x1f80000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0x900a7720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0x900ab720 | 0x1f80000 |
260
+ | 157 | blk.17.attn_k.weight | 0x9202b720 | 0x1b8000 |
261
+ | 158 | blk.17.attn_norm.weight | 0x921e3720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0x921e7720 | 0x900000 |
263
+ | 160 | blk.17.attn_q.weight | 0x92ae7720 | 0x6e0000 |
264
+ | 161 | blk.17.attn_v.weight | 0x931c7720 | 0x220000 |
265
+ | 162 | blk.17.ffn_down.weight | 0x933e7720 | 0x2680000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0x95a67720 | 0x1f80000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0x979e7720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0x979eb720 | 0x1f80000 |
269
+ | 166 | blk.18.attn_k.weight | 0x9996b720 | 0x240000 |
270
+ | 167 | blk.18.attn_norm.weight | 0x99bab720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0x99baf720 | 0x900000 |
272
+ | 169 | blk.18.attn_q.weight | 0x9a4af720 | 0x900000 |
273
+ | 170 | blk.18.attn_v.weight | 0x9adaf720 | 0x240000 |
274
+ | 171 | blk.18.ffn_down.weight | 0x9afef720 | 0x2680000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0x9d66f720 | 0x1f80000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0x9f5ef720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0x9f5f3720 | 0x1f80000 |
278
+ | 175 | blk.19.attn_k.weight | 0xa1573720 | 0x240000 |
279
+ | 176 | blk.19.attn_norm.weight | 0xa17b3720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0xa17b7720 | 0x900000 |
281
+ | 178 | blk.19.attn_q.weight | 0xa20b7720 | 0x900000 |
282
+ | 179 | blk.19.attn_v.weight | 0xa29b7720 | 0x240000 |
283
+ | 180 | blk.19.ffn_down.weight | 0xa2bf7720 | 0x2680000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0xa5277720 | 0x1f80000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0xa71f7720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0xa71fb720 | 0x1f80000 |
287
+ | 184 | blk.20.attn_k.weight | 0xa917b720 | 0x240000 |
288
+ | 185 | blk.20.attn_norm.weight | 0xa93bb720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0xa93bf720 | 0x900000 |
290
+ | 187 | blk.20.attn_q.weight | 0xa9cbf720 | 0x900000 |
291
+ | 188 | blk.20.attn_v.weight | 0xaa5bf720 | 0x240000 |
292
+ | 189 | blk.20.ffn_down.weight | 0xaa7ff720 | 0x2680000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0xace7f720 | 0x1f80000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0xaedff720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0xaee03720 | 0x1f80000 |
296
+ | 193 | blk.21.attn_k.weight | 0xb0d83720 | 0x240000 |
297
+ | 194 | blk.21.attn_norm.weight | 0xb0fc3720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0xb0fc7720 | 0x900000 |
299
+ | 196 | blk.21.attn_q.weight | 0xb18c7720 | 0x900000 |
300
+ | 197 | blk.21.attn_v.weight | 0xb21c7720 | 0x240000 |
301
+ | 198 | blk.21.ffn_down.weight | 0xb2407720 | 0x2680000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0xb4a87720 | 0x1f80000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0xb6a07720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0xb6a0b720 | 0x1f80000 |
305
+ | 202 | blk.22.attn_k.weight | 0xb898b720 | 0x240000 |
306
+ | 203 | blk.22.attn_norm.weight | 0xb8bcb720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0xb8bcf720 | 0x900000 |
308
+ | 205 | blk.22.attn_q.weight | 0xb94cf720 | 0x900000 |
309
+ | 206 | blk.22.attn_v.weight | 0xb9dcf720 | 0x240000 |
310
+ | 207 | blk.22.ffn_down.weight | 0xba00f720 | 0x2680000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0xbc68f720 | 0x1f80000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0xbe60f720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0xbe613720 | 0x1f80000 |
314
+ | 211 | blk.23.attn_k.weight | 0xc0593720 | 0x240000 |
315
+ | 212 | blk.23.attn_norm.weight | 0xc07d3720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0xc07d7720 | 0x900000 |
317
+ | 214 | blk.23.attn_q.weight | 0xc10d7720 | 0x900000 |
318
+ | 215 | blk.23.attn_v.weight | 0xc19d7720 | 0x240000 |
319
+ | 216 | blk.23.ffn_down.weight | 0xc1c17720 | 0x2680000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0xc4297720 | 0x1f80000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0xc6217720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0xc621b720 | 0x1f80000 |
323
+ | 220 | blk.24.attn_k.weight | 0xc819b720 | 0x240000 |
324
+ | 221 | blk.24.attn_norm.weight | 0xc83db720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0xc83df720 | 0x900000 |
326
+ | 223 | blk.24.attn_q.weight | 0xc8cdf720 | 0x900000 |
327
+ | 224 | blk.24.attn_v.weight | 0xc95df720 | 0x240000 |
328
+ | 225 | blk.24.ffn_down.weight | 0xc981f720 | 0x2680000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0xcbe9f720 | 0x1f80000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0xcde1f720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0xcde23720 | 0x1f80000 |
332
+ | 229 | blk.25.attn_k.weight | 0xcfda3720 | 0x240000 |
333
+ | 230 | blk.25.attn_norm.weight | 0xcffe3720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0xcffe7720 | 0x900000 |
335
+ | 232 | blk.25.attn_q.weight | 0xd08e7720 | 0x900000 |
336
+ | 233 | blk.25.attn_v.weight | 0xd11e7720 | 0x240000 |
337
+ | 234 | blk.25.ffn_down.weight | 0xd1427720 | 0x2680000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0xd3aa7720 | 0x1f80000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0xd5a27720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0xd5a2b720 | 0x1f80000 |
341
+ | 238 | blk.26.attn_k.weight | 0xd79ab720 | 0x240000 |
342
+ | 239 | blk.26.attn_norm.weight | 0xd7beb720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0xd7bef720 | 0x900000 |
344
+ | 241 | blk.26.attn_q.weight | 0xd84ef720 | 0x900000 |
345
+ | 242 | blk.26.attn_v.weight | 0xd8def720 | 0x240000 |
346
+ | 243 | blk.26.ffn_down.weight | 0xd902f720 | 0x2680000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0xdb6af720 | 0x1f80000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0xdd62f720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0xdd633720 | 0x1f80000 |
350
+ | 247 | blk.27.attn_k.weight | 0xdf5b3720 | 0x240000 |
351
+ | 248 | blk.27.attn_norm.weight | 0xdf7f3720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0xdf7f7720 | 0x900000 |
353
+ | 250 | blk.27.attn_q.weight | 0xe00f7720 | 0x900000 |
354
+ | 251 | blk.27.attn_v.weight | 0xe09f7720 | 0x240000 |
355
+ | 252 | blk.27.ffn_down.weight | 0xe0c37720 | 0x2680000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0xe32b7720 | 0x1f80000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0xe5237720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0xe523b720 | 0x1f80000 |
359
+ | 256 | blk.28.attn_k.weight | 0xe71bb720 | 0x240000 |
360
+ | 257 | blk.28.attn_norm.weight | 0xe73fb720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0xe73ff720 | 0x900000 |
362
+ | 259 | blk.28.attn_q.weight | 0xe7cff720 | 0x900000 |
363
+ | 260 | blk.28.attn_v.weight | 0xe85ff720 | 0x240000 |
364
+ | 261 | blk.28.ffn_down.weight | 0xe883f720 | 0x2680000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0xeaebf720 | 0x1f80000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0xece3f720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0xece43720 | 0x1f80000 |
368
+ | 265 | blk.29.attn_k.weight | 0xeedc3720 | 0x240000 |
369
+ | 266 | blk.29.attn_norm.weight | 0xef003720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0xef007720 | 0x900000 |
371
+ | 268 | blk.29.attn_q.weight | 0xef907720 | 0x900000 |
372
+ | 269 | blk.29.attn_v.weight | 0xf0207720 | 0x240000 |
373
+ | 270 | blk.29.ffn_down.weight | 0xf0447720 | 0x2680000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0xf2ac7720 | 0x1f80000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0xf4a47720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0xf4a4b720 | 0x1f80000 |
377
+ | 274 | blk.30.attn_k.weight | 0xf69cb720 | 0x240000 |
378
+ | 275 | blk.30.attn_norm.weight | 0xf6c0b720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0xf6c0f720 | 0x900000 |
380
+ | 277 | blk.30.attn_q.weight | 0xf750f720 | 0x900000 |
381
+ | 278 | blk.30.attn_v.weight | 0xf7e0f720 | 0x240000 |
382
+ | 279 | blk.30.ffn_down.weight | 0xf804f720 | 0x2680000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0xfa6cf720 | 0x1f80000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0xfc64f720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0xfc653720 | 0x1f80000 |
386
+ | 283 | blk.31.attn_k.weight | 0xfe5d3720 | 0x1b8000 |
387
+ | 284 | blk.31.attn_norm.weight | 0xfe78b720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0xfe78f720 | 0x900000 |
389
+ | 286 | blk.31.attn_q.weight | 0xff08f720 | 0x6e0000 |
390
+ | 287 | blk.31.attn_v.weight | 0xff76f720 | 0x220000 |
391
+ | 288 | blk.31.ffn_down.weight | 0xff98f720 | 0x2680000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0x10200f720 | 0x1f80000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0x103f8f720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0x103f93720 | 0x1f80000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-------|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | IQ4_NL |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | IQ3_S |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-------|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-------|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-------|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-------|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-------|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-------|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-------|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-------|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-------|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-------|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ3_S |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_NL |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-------|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ3_S |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ4_NL |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | IQ3_S |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | IQ4_XS |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | IQ4_NL |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-Q3_K_L.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-Q3_K_L.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 13 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-Q3\_K\_L.gguf - GGUF Internal File Dump](#watt-tool-8b-q3_k_lgguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0xd746000 |
104
+ | 1 | output_norm.weight | 0xdebf620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0xdec3620 | 0x100 |
106
+ | 3 | token_embd.weight | 0xdec3720 | 0xd746000 |
107
+ | 4 | blk.0.attn_k.weight | 0x1b609720 | 0x150000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x1b759720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x1b75d720 | 0xb00000 |
110
+ | 7 | blk.0.attn_q.weight | 0x1c25d720 | 0x540000 |
111
+ | 8 | blk.0.attn_v.weight | 0x1c79d720 | 0x240000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x1c9dd720 | 0x1f80000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x1e95d720 | 0x1260000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x1fbbd720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x1fbc1720 | 0x1260000 |
116
+ | 13 | blk.1.attn_k.weight | 0x20e21720 | 0x150000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x20f71720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x20f75720 | 0xb00000 |
119
+ | 16 | blk.1.attn_q.weight | 0x21a75720 | 0x540000 |
120
+ | 17 | blk.1.attn_v.weight | 0x21fb5720 | 0x240000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x221f5720 | 0x2680000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x24875720 | 0x1260000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x25ad5720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x25ad9720 | 0x1260000 |
125
+ | 22 | blk.2.attn_k.weight | 0x26d39720 | 0x150000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x26e89720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x26e8d720 | 0xb00000 |
128
+ | 25 | blk.2.attn_q.weight | 0x2798d720 | 0x540000 |
129
+ | 26 | blk.2.attn_v.weight | 0x27ecd720 | 0x240000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x2810d720 | 0x1f80000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x2a08d720 | 0x1260000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x2b2ed720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x2b2f1720 | 0x1260000 |
134
+ | 31 | blk.3.attn_k.weight | 0x2c551720 | 0x150000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x2c6a1720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x2c6a5720 | 0xb00000 |
137
+ | 34 | blk.3.attn_q.weight | 0x2d1a5720 | 0x540000 |
138
+ | 35 | blk.3.attn_v.weight | 0x2d6e5720 | 0x240000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x2d925720 | 0x1f80000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x2f8a5720 | 0x1260000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x30b05720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x30b09720 | 0x1260000 |
143
+ | 40 | blk.4.attn_k.weight | 0x31d69720 | 0x150000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x31eb9720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x31ebd720 | 0xb00000 |
146
+ | 43 | blk.4.attn_q.weight | 0x329bd720 | 0x540000 |
147
+ | 44 | blk.4.attn_v.weight | 0x32efd720 | 0x240000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x3313d720 | 0x1f80000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x350bd720 | 0x1260000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x3631d720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x36321720 | 0x1260000 |
152
+ | 49 | blk.5.attn_k.weight | 0x37581720 | 0x150000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x376d1720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x376d5720 | 0xb00000 |
155
+ | 52 | blk.5.attn_q.weight | 0x381d5720 | 0x540000 |
156
+ | 53 | blk.5.attn_v.weight | 0x38715720 | 0x240000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x38955720 | 0x1f80000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x3a8d5720 | 0x1260000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x3bb35720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x3bb39720 | 0x1260000 |
161
+ | 58 | blk.6.attn_k.weight | 0x3cd99720 | 0x150000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x3cee9720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x3ceed720 | 0xb00000 |
164
+ | 61 | blk.6.attn_q.weight | 0x3d9ed720 | 0x540000 |
165
+ | 62 | blk.6.attn_v.weight | 0x3df2d720 | 0x240000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x3e16d720 | 0x1f80000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x400ed720 | 0x1260000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x4134d720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x41351720 | 0x1260000 |
170
+ | 67 | blk.7.attn_k.weight | 0x425b1720 | 0x150000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x42701720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x42705720 | 0xb00000 |
173
+ | 70 | blk.7.attn_q.weight | 0x43205720 | 0x540000 |
174
+ | 71 | blk.7.attn_v.weight | 0x43745720 | 0x240000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x43985720 | 0x1f80000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x45905720 | 0x1260000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x46b65720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x46b69720 | 0x1260000 |
179
+ | 76 | blk.8.attn_k.weight | 0x47dc9720 | 0x150000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x47f19720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x47f1d720 | 0xb00000 |
182
+ | 79 | blk.8.attn_q.weight | 0x48a1d720 | 0x540000 |
183
+ | 80 | blk.8.attn_v.weight | 0x48f5d720 | 0x240000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x4919d720 | 0x1f80000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x4b11d720 | 0x1260000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x4c37d720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x4c381720 | 0x1260000 |
188
+ | 85 | blk.9.attn_k.weight | 0x4d5e1720 | 0x150000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x4d731720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x4d735720 | 0xb00000 |
191
+ | 88 | blk.9.attn_q.weight | 0x4e235720 | 0x540000 |
192
+ | 89 | blk.9.attn_v.weight | 0x4e775720 | 0x240000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x4e9b5720 | 0x1f80000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x50935720 | 0x1260000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0x51b95720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0x51b99720 | 0x1260000 |
197
+ | 94 | blk.10.attn_k.weight | 0x52df9720 | 0x150000 |
198
+ | 95 | blk.10.attn_norm.weight | 0x52f49720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0x52f4d720 | 0xb00000 |
200
+ | 97 | blk.10.attn_q.weight | 0x53a4d720 | 0x540000 |
201
+ | 98 | blk.10.attn_v.weight | 0x53f8d720 | 0x240000 |
202
+ | 99 | blk.10.ffn_down.weight | 0x541cd720 | 0x1f80000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0x5614d720 | 0x1260000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0x573ad720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0x573b1720 | 0x1260000 |
206
+ | 103 | blk.11.attn_k.weight | 0x58611720 | 0x150000 |
207
+ | 104 | blk.11.attn_norm.weight | 0x58761720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0x58765720 | 0xb00000 |
209
+ | 106 | blk.11.attn_q.weight | 0x59265720 | 0x540000 |
210
+ | 107 | blk.11.attn_v.weight | 0x597a5720 | 0x240000 |
211
+ | 108 | blk.11.ffn_down.weight | 0x599e5720 | 0x1f80000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0x5b965720 | 0x1260000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0x5cbc5720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0x5cbc9720 | 0x1260000 |
215
+ | 112 | blk.12.attn_k.weight | 0x5de29720 | 0x150000 |
216
+ | 113 | blk.12.attn_norm.weight | 0x5df79720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0x5df7d720 | 0xb00000 |
218
+ | 115 | blk.12.attn_q.weight | 0x5ea7d720 | 0x540000 |
219
+ | 116 | blk.12.attn_v.weight | 0x5efbd720 | 0x240000 |
220
+ | 117 | blk.12.ffn_down.weight | 0x5f1fd720 | 0x1f80000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0x6117d720 | 0x1260000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0x623dd720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0x623e1720 | 0x1260000 |
224
+ | 121 | blk.13.attn_k.weight | 0x63641720 | 0x1b8000 |
225
+ | 122 | blk.13.attn_norm.weight | 0x637f9720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0x637fd720 | 0xb00000 |
227
+ | 124 | blk.13.attn_q.weight | 0x642fd720 | 0x6e0000 |
228
+ | 125 | blk.13.attn_v.weight | 0x649dd720 | 0x240000 |
229
+ | 126 | blk.13.ffn_down.weight | 0x64c1d720 | 0x1f80000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0x66b9d720 | 0x1260000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0x67dfd720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0x67e01720 | 0x1260000 |
233
+ | 130 | blk.14.attn_k.weight | 0x69061720 | 0x1b8000 |
234
+ | 131 | blk.14.attn_norm.weight | 0x69219720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0x6921d720 | 0xb00000 |
236
+ | 133 | blk.14.attn_q.weight | 0x69d1d720 | 0x6e0000 |
237
+ | 134 | blk.14.attn_v.weight | 0x6a3fd720 | 0x240000 |
238
+ | 135 | blk.14.ffn_down.weight | 0x6a63d720 | 0x1f80000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0x6c5bd720 | 0x1260000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0x6d81d720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0x6d821720 | 0x1260000 |
242
+ | 139 | blk.15.attn_k.weight | 0x6ea81720 | 0x150000 |
243
+ | 140 | blk.15.attn_norm.weight | 0x6ebd1720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0x6ebd5720 | 0xb00000 |
245
+ | 142 | blk.15.attn_q.weight | 0x6f6d5720 | 0x540000 |
246
+ | 143 | blk.15.attn_v.weight | 0x6fc15720 | 0x240000 |
247
+ | 144 | blk.15.ffn_down.weight | 0x6fe55720 | 0x1f80000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0x71dd5720 | 0x1260000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0x73035720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0x73039720 | 0x1260000 |
251
+ | 148 | blk.16.attn_k.weight | 0x74299720 | 0x1b8000 |
252
+ | 149 | blk.16.attn_norm.weight | 0x74451720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0x74455720 | 0xb00000 |
254
+ | 151 | blk.16.attn_q.weight | 0x74f55720 | 0x6e0000 |
255
+ | 152 | blk.16.attn_v.weight | 0x75635720 | 0x240000 |
256
+ | 153 | blk.16.ffn_down.weight | 0x75875720 | 0x1f80000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0x777f5720 | 0x1810000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0x79005720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0x79009720 | 0x1810000 |
260
+ | 157 | blk.17.attn_k.weight | 0x7a819720 | 0x150000 |
261
+ | 158 | blk.17.attn_norm.weight | 0x7a969720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0x7a96d720 | 0xb00000 |
263
+ | 160 | blk.17.attn_q.weight | 0x7b46d720 | 0x540000 |
264
+ | 161 | blk.17.attn_v.weight | 0x7b9ad720 | 0x240000 |
265
+ | 162 | blk.17.ffn_down.weight | 0x7bbed720 | 0x2680000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0x7e26d720 | 0x1810000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0x7fa7d720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0x7fa81720 | 0x1810000 |
269
+ | 166 | blk.18.attn_k.weight | 0x81291720 | 0x1b8000 |
270
+ | 167 | blk.18.attn_norm.weight | 0x81449720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0x8144d720 | 0xb00000 |
272
+ | 169 | blk.18.attn_q.weight | 0x81f4d720 | 0x6e0000 |
273
+ | 170 | blk.18.attn_v.weight | 0x8262d720 | 0x240000 |
274
+ | 171 | blk.18.ffn_down.weight | 0x8286d720 | 0x2680000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0x84eed720 | 0x1810000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0x866fd720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0x86701720 | 0x1810000 |
278
+ | 175 | blk.19.attn_k.weight | 0x87f11720 | 0x1b8000 |
279
+ | 176 | blk.19.attn_norm.weight | 0x880c9720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0x880cd720 | 0xb00000 |
281
+ | 178 | blk.19.attn_q.weight | 0x88bcd720 | 0x6e0000 |
282
+ | 179 | blk.19.attn_v.weight | 0x892ad720 | 0x240000 |
283
+ | 180 | blk.19.ffn_down.weight | 0x894ed720 | 0x2680000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0x8bb6d720 | 0x1810000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0x8d37d720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0x8d381720 | 0x1810000 |
287
+ | 184 | blk.20.attn_k.weight | 0x8eb91720 | 0x1b8000 |
288
+ | 185 | blk.20.attn_norm.weight | 0x8ed49720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0x8ed4d720 | 0xb00000 |
290
+ | 187 | blk.20.attn_q.weight | 0x8f84d720 | 0x6e0000 |
291
+ | 188 | blk.20.attn_v.weight | 0x8ff2d720 | 0x240000 |
292
+ | 189 | blk.20.ffn_down.weight | 0x9016d720 | 0x2680000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0x927ed720 | 0x1810000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0x93ffd720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0x94001720 | 0x1810000 |
296
+ | 193 | blk.21.attn_k.weight | 0x95811720 | 0x1b8000 |
297
+ | 194 | blk.21.attn_norm.weight | 0x959c9720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0x959cd720 | 0xb00000 |
299
+ | 196 | blk.21.attn_q.weight | 0x964cd720 | 0x6e0000 |
300
+ | 197 | blk.21.attn_v.weight | 0x96bad720 | 0x240000 |
301
+ | 198 | blk.21.ffn_down.weight | 0x96ded720 | 0x2680000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0x9946d720 | 0x1810000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0x9ac7d720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0x9ac81720 | 0x1810000 |
305
+ | 202 | blk.22.attn_k.weight | 0x9c491720 | 0x1b8000 |
306
+ | 203 | blk.22.attn_norm.weight | 0x9c649720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0x9c64d720 | 0xb00000 |
308
+ | 205 | blk.22.attn_q.weight | 0x9d14d720 | 0x6e0000 |
309
+ | 206 | blk.22.attn_v.weight | 0x9d82d720 | 0x240000 |
310
+ | 207 | blk.22.ffn_down.weight | 0x9da6d720 | 0x2680000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0xa00ed720 | 0x1810000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0xa18fd720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0xa1901720 | 0x1810000 |
314
+ | 211 | blk.23.attn_k.weight | 0xa3111720 | 0x1b8000 |
315
+ | 212 | blk.23.attn_norm.weight | 0xa32c9720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0xa32cd720 | 0xb00000 |
317
+ | 214 | blk.23.attn_q.weight | 0xa3dcd720 | 0x6e0000 |
318
+ | 215 | blk.23.attn_v.weight | 0xa44ad720 | 0x240000 |
319
+ | 216 | blk.23.ffn_down.weight | 0xa46ed720 | 0x2680000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0xa6d6d720 | 0x1810000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0xa857d720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0xa8581720 | 0x1810000 |
323
+ | 220 | blk.24.attn_k.weight | 0xa9d91720 | 0x1b8000 |
324
+ | 221 | blk.24.attn_norm.weight | 0xa9f49720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0xa9f4d720 | 0xb00000 |
326
+ | 223 | blk.24.attn_q.weight | 0xaaa4d720 | 0x6e0000 |
327
+ | 224 | blk.24.attn_v.weight | 0xab12d720 | 0x240000 |
328
+ | 225 | blk.24.ffn_down.weight | 0xab36d720 | 0x2680000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0xad9ed720 | 0x1810000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0xaf1fd720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0xaf201720 | 0x1810000 |
332
+ | 229 | blk.25.attn_k.weight | 0xb0a11720 | 0x1b8000 |
333
+ | 230 | blk.25.attn_norm.weight | 0xb0bc9720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0xb0bcd720 | 0xb00000 |
335
+ | 232 | blk.25.attn_q.weight | 0xb16cd720 | 0x6e0000 |
336
+ | 233 | blk.25.attn_v.weight | 0xb1dad720 | 0x240000 |
337
+ | 234 | blk.25.ffn_down.weight | 0xb1fed720 | 0x2680000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0xb466d720 | 0x1810000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0xb5e7d720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0xb5e81720 | 0x1810000 |
341
+ | 238 | blk.26.attn_k.weight | 0xb7691720 | 0x1b8000 |
342
+ | 239 | blk.26.attn_norm.weight | 0xb7849720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0xb784d720 | 0xb00000 |
344
+ | 241 | blk.26.attn_q.weight | 0xb834d720 | 0x6e0000 |
345
+ | 242 | blk.26.attn_v.weight | 0xb8a2d720 | 0x240000 |
346
+ | 243 | blk.26.ffn_down.weight | 0xb8c6d720 | 0x2680000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0xbb2ed720 | 0x1810000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0xbcafd720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0xbcb01720 | 0x1810000 |
350
+ | 247 | blk.27.attn_k.weight | 0xbe311720 | 0x1b8000 |
351
+ | 248 | blk.27.attn_norm.weight | 0xbe4c9720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0xbe4cd720 | 0xb00000 |
353
+ | 250 | blk.27.attn_q.weight | 0xbefcd720 | 0x6e0000 |
354
+ | 251 | blk.27.attn_v.weight | 0xbf6ad720 | 0x240000 |
355
+ | 252 | blk.27.ffn_down.weight | 0xbf8ed720 | 0x2680000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0xc1f6d720 | 0x1810000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0xc377d720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0xc3781720 | 0x1810000 |
359
+ | 256 | blk.28.attn_k.weight | 0xc4f91720 | 0x1b8000 |
360
+ | 257 | blk.28.attn_norm.weight | 0xc5149720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0xc514d720 | 0xb00000 |
362
+ | 259 | blk.28.attn_q.weight | 0xc5c4d720 | 0x6e0000 |
363
+ | 260 | blk.28.attn_v.weight | 0xc632d720 | 0x240000 |
364
+ | 261 | blk.28.ffn_down.weight | 0xc656d720 | 0x2680000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0xc8bed720 | 0x1810000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0xca3fd720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0xca401720 | 0x1810000 |
368
+ | 265 | blk.29.attn_k.weight | 0xcbc11720 | 0x1b8000 |
369
+ | 266 | blk.29.attn_norm.weight | 0xcbdc9720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0xcbdcd720 | 0xb00000 |
371
+ | 268 | blk.29.attn_q.weight | 0xcc8cd720 | 0x6e0000 |
372
+ | 269 | blk.29.attn_v.weight | 0xccfad720 | 0x240000 |
373
+ | 270 | blk.29.ffn_down.weight | 0xcd1ed720 | 0x2680000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0xcf86d720 | 0x1810000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0xd107d720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0xd1081720 | 0x1810000 |
377
+ | 274 | blk.30.attn_k.weight | 0xd2891720 | 0x1b8000 |
378
+ | 275 | blk.30.attn_norm.weight | 0xd2a49720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0xd2a4d720 | 0xb00000 |
380
+ | 277 | blk.30.attn_q.weight | 0xd354d720 | 0x6e0000 |
381
+ | 278 | blk.30.attn_v.weight | 0xd3c2d720 | 0x240000 |
382
+ | 279 | blk.30.ffn_down.weight | 0xd3e6d720 | 0x2680000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0xd64ed720 | 0x1810000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0xd7cfd720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0xd7d01720 | 0x1810000 |
386
+ | 283 | blk.31.attn_k.weight | 0xd9511720 | 0x150000 |
387
+ | 284 | blk.31.attn_norm.weight | 0xd9661720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0xd9665720 | 0xb00000 |
389
+ | 286 | blk.31.attn_q.weight | 0xda165720 | 0x540000 |
390
+ | 287 | blk.31.attn_v.weight | 0xda6a5720 | 0x240000 |
391
+ | 288 | blk.31.ffn_down.weight | 0xda8e5720 | 0x2680000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0xdcf65720 | 0x1810000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0xde775720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0xde779720 | 0x1810000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q3_K |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q3_K |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-Q3_K_M.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-Q3_K_M.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 12 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-Q3\_K\_M.gguf - GGUF Internal File Dump](#watt-tool-8b-q3_k_mgguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0xd746000 |
104
+ | 1 | output_norm.weight | 0xdebf620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0xdec3620 | 0x100 |
106
+ | 3 | token_embd.weight | 0xdec3720 | 0xd746000 |
107
+ | 4 | blk.0.attn_k.weight | 0x1b609720 | 0x150000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x1b759720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x1b75d720 | 0x900000 |
110
+ | 7 | blk.0.attn_q.weight | 0x1c05d720 | 0x540000 |
111
+ | 8 | blk.0.attn_v.weight | 0x1c59d720 | 0x1b8000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x1c755720 | 0x1f80000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x1e6d5720 | 0x1260000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x1f935720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x1f939720 | 0x1260000 |
116
+ | 13 | blk.1.attn_k.weight | 0x20b99720 | 0x150000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x20ce9720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x20ced720 | 0x900000 |
119
+ | 16 | blk.1.attn_q.weight | 0x215ed720 | 0x540000 |
120
+ | 17 | blk.1.attn_v.weight | 0x21b2d720 | 0x1b8000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x21ce5720 | 0x1f80000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x23c65720 | 0x1260000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x24ec5720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x24ec9720 | 0x1260000 |
125
+ | 22 | blk.2.attn_k.weight | 0x26129720 | 0x150000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x26279720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x2627d720 | 0x900000 |
128
+ | 25 | blk.2.attn_q.weight | 0x26b7d720 | 0x540000 |
129
+ | 26 | blk.2.attn_v.weight | 0x270bd720 | 0x1b8000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x27275720 | 0x1f80000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x291f5720 | 0x1260000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x2a455720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x2a459720 | 0x1260000 |
134
+ | 31 | blk.3.attn_k.weight | 0x2b6b9720 | 0x150000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x2b809720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x2b80d720 | 0x900000 |
137
+ | 34 | blk.3.attn_q.weight | 0x2c10d720 | 0x540000 |
138
+ | 35 | blk.3.attn_v.weight | 0x2c64d720 | 0x1b8000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x2c805720 | 0x1f80000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x2e785720 | 0x1260000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x2f9e5720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x2f9e9720 | 0x1260000 |
143
+ | 40 | blk.4.attn_k.weight | 0x30c49720 | 0x150000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x30d99720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x30d9d720 | 0x900000 |
146
+ | 43 | blk.4.attn_q.weight | 0x3169d720 | 0x540000 |
147
+ | 44 | blk.4.attn_v.weight | 0x31bdd720 | 0x1b8000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x31d95720 | 0x1f80000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x33d15720 | 0x1260000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x34f75720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x34f79720 | 0x1260000 |
152
+ | 49 | blk.5.attn_k.weight | 0x361d9720 | 0x150000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x36329720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x3632d720 | 0x900000 |
155
+ | 52 | blk.5.attn_q.weight | 0x36c2d720 | 0x540000 |
156
+ | 53 | blk.5.attn_v.weight | 0x3716d720 | 0x1b8000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x37325720 | 0x1f80000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x392a5720 | 0x1260000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x3a505720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x3a509720 | 0x1260000 |
161
+ | 58 | blk.6.attn_k.weight | 0x3b769720 | 0x150000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x3b8b9720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x3b8bd720 | 0x900000 |
164
+ | 61 | blk.6.attn_q.weight | 0x3c1bd720 | 0x540000 |
165
+ | 62 | blk.6.attn_v.weight | 0x3c6fd720 | 0x1b8000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x3c8b5720 | 0x1f80000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x3e835720 | 0x1260000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x3fa95720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x3fa99720 | 0x1260000 |
170
+ | 67 | blk.7.attn_k.weight | 0x40cf9720 | 0x150000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x40e49720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x40e4d720 | 0x900000 |
173
+ | 70 | blk.7.attn_q.weight | 0x4174d720 | 0x540000 |
174
+ | 71 | blk.7.attn_v.weight | 0x41c8d720 | 0x1b8000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x41e45720 | 0x1f80000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x43dc5720 | 0x1260000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x45025720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x45029720 | 0x1260000 |
179
+ | 76 | blk.8.attn_k.weight | 0x46289720 | 0x150000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x463d9720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x463dd720 | 0x900000 |
182
+ | 79 | blk.8.attn_q.weight | 0x46cdd720 | 0x540000 |
183
+ | 80 | blk.8.attn_v.weight | 0x4721d720 | 0x1b8000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x473d5720 | 0x1f80000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x49355720 | 0x1260000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x4a5b5720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x4a5b9720 | 0x1260000 |
188
+ | 85 | blk.9.attn_k.weight | 0x4b819720 | 0x150000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x4b969720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x4b96d720 | 0x900000 |
191
+ | 88 | blk.9.attn_q.weight | 0x4c26d720 | 0x540000 |
192
+ | 89 | blk.9.attn_v.weight | 0x4c7ad720 | 0x1b8000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x4c965720 | 0x1f80000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x4e8e5720 | 0x1260000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0x4fb45720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0x4fb49720 | 0x1260000 |
197
+ | 94 | blk.10.attn_k.weight | 0x50da9720 | 0x150000 |
198
+ | 95 | blk.10.attn_norm.weight | 0x50ef9720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0x50efd720 | 0x900000 |
200
+ | 97 | blk.10.attn_q.weight | 0x517fd720 | 0x540000 |
201
+ | 98 | blk.10.attn_v.weight | 0x51d3d720 | 0x1b8000 |
202
+ | 99 | blk.10.ffn_down.weight | 0x51ef5720 | 0x1f80000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0x53e75720 | 0x1260000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0x550d5720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0x550d9720 | 0x1260000 |
206
+ | 103 | blk.11.attn_k.weight | 0x56339720 | 0x150000 |
207
+ | 104 | blk.11.attn_norm.weight | 0x56489720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0x5648d720 | 0x900000 |
209
+ | 106 | blk.11.attn_q.weight | 0x56d8d720 | 0x540000 |
210
+ | 107 | blk.11.attn_v.weight | 0x572cd720 | 0x1b8000 |
211
+ | 108 | blk.11.ffn_down.weight | 0x57485720 | 0x1f80000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0x59405720 | 0x1260000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0x5a665720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0x5a669720 | 0x1260000 |
215
+ | 112 | blk.12.attn_k.weight | 0x5b8c9720 | 0x150000 |
216
+ | 113 | blk.12.attn_norm.weight | 0x5ba19720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0x5ba1d720 | 0x900000 |
218
+ | 115 | blk.12.attn_q.weight | 0x5c31d720 | 0x540000 |
219
+ | 116 | blk.12.attn_v.weight | 0x5c85d720 | 0x1b8000 |
220
+ | 117 | blk.12.ffn_down.weight | 0x5ca15720 | 0x1f80000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0x5e995720 | 0x1260000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0x5fbf5720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0x5fbf9720 | 0x1260000 |
224
+ | 121 | blk.13.attn_k.weight | 0x60e59720 | 0x1b8000 |
225
+ | 122 | blk.13.attn_norm.weight | 0x61011720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0x61015720 | 0x900000 |
227
+ | 124 | blk.13.attn_q.weight | 0x61915720 | 0x6e0000 |
228
+ | 125 | blk.13.attn_v.weight | 0x61ff5720 | 0x240000 |
229
+ | 126 | blk.13.ffn_down.weight | 0x62235720 | 0x1f80000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0x641b5720 | 0x1260000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0x65415720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0x65419720 | 0x1260000 |
233
+ | 130 | blk.14.attn_k.weight | 0x66679720 | 0x1b8000 |
234
+ | 131 | blk.14.attn_norm.weight | 0x66831720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0x66835720 | 0x900000 |
236
+ | 133 | blk.14.attn_q.weight | 0x67135720 | 0x6e0000 |
237
+ | 134 | blk.14.attn_v.weight | 0x67815720 | 0x240000 |
238
+ | 135 | blk.14.ffn_down.weight | 0x67a55720 | 0x1f80000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0x699d5720 | 0x1260000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0x6ac35720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0x6ac39720 | 0x1260000 |
242
+ | 139 | blk.15.attn_k.weight | 0x6be99720 | 0x150000 |
243
+ | 140 | blk.15.attn_norm.weight | 0x6bfe9720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0x6bfed720 | 0x900000 |
245
+ | 142 | blk.15.attn_q.weight | 0x6c8ed720 | 0x540000 |
246
+ | 143 | blk.15.attn_v.weight | 0x6ce2d720 | 0x1b8000 |
247
+ | 144 | blk.15.ffn_down.weight | 0x6cfe5720 | 0x1f80000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0x6ef65720 | 0x1260000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0x701c5720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0x701c9720 | 0x1260000 |
251
+ | 148 | blk.16.attn_k.weight | 0x71429720 | 0x1b8000 |
252
+ | 149 | blk.16.attn_norm.weight | 0x715e1720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0x715e5720 | 0x900000 |
254
+ | 151 | blk.16.attn_q.weight | 0x71ee5720 | 0x6e0000 |
255
+ | 152 | blk.16.attn_v.weight | 0x725c5720 | 0x240000 |
256
+ | 153 | blk.16.ffn_down.weight | 0x72805720 | 0x1f80000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0x74785720 | 0x1810000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0x75f95720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0x75f99720 | 0x1810000 |
260
+ | 157 | blk.17.attn_k.weight | 0x777a9720 | 0x150000 |
261
+ | 158 | blk.17.attn_norm.weight | 0x778f9720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0x778fd720 | 0x900000 |
263
+ | 160 | blk.17.attn_q.weight | 0x781fd720 | 0x540000 |
264
+ | 161 | blk.17.attn_v.weight | 0x7873d720 | 0x1b8000 |
265
+ | 162 | blk.17.ffn_down.weight | 0x788f5720 | 0x1f80000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0x7a875720 | 0x1810000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0x7c085720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0x7c089720 | 0x1810000 |
269
+ | 166 | blk.18.attn_k.weight | 0x7d899720 | 0x1b8000 |
270
+ | 167 | blk.18.attn_norm.weight | 0x7da51720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0x7da55720 | 0x900000 |
272
+ | 169 | blk.18.attn_q.weight | 0x7e355720 | 0x6e0000 |
273
+ | 170 | blk.18.attn_v.weight | 0x7ea35720 | 0x240000 |
274
+ | 171 | blk.18.ffn_down.weight | 0x7ec75720 | 0x1f80000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0x80bf5720 | 0x1810000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0x82405720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0x82409720 | 0x1810000 |
278
+ | 175 | blk.19.attn_k.weight | 0x83c19720 | 0x1b8000 |
279
+ | 176 | blk.19.attn_norm.weight | 0x83dd1720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0x83dd5720 | 0x900000 |
281
+ | 178 | blk.19.attn_q.weight | 0x846d5720 | 0x6e0000 |
282
+ | 179 | blk.19.attn_v.weight | 0x84db5720 | 0x240000 |
283
+ | 180 | blk.19.ffn_down.weight | 0x84ff5720 | 0x1f80000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0x86f75720 | 0x1810000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0x88785720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0x88789720 | 0x1810000 |
287
+ | 184 | blk.20.attn_k.weight | 0x89f99720 | 0x1b8000 |
288
+ | 185 | blk.20.attn_norm.weight | 0x8a151720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0x8a155720 | 0x900000 |
290
+ | 187 | blk.20.attn_q.weight | 0x8aa55720 | 0x6e0000 |
291
+ | 188 | blk.20.attn_v.weight | 0x8b135720 | 0x240000 |
292
+ | 189 | blk.20.ffn_down.weight | 0x8b375720 | 0x1f80000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0x8d2f5720 | 0x1810000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0x8eb05720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0x8eb09720 | 0x1810000 |
296
+ | 193 | blk.21.attn_k.weight | 0x90319720 | 0x1b8000 |
297
+ | 194 | blk.21.attn_norm.weight | 0x904d1720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0x904d5720 | 0x900000 |
299
+ | 196 | blk.21.attn_q.weight | 0x90dd5720 | 0x6e0000 |
300
+ | 197 | blk.21.attn_v.weight | 0x914b5720 | 0x240000 |
301
+ | 198 | blk.21.ffn_down.weight | 0x916f5720 | 0x1f80000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0x93675720 | 0x1810000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0x94e85720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0x94e89720 | 0x1810000 |
305
+ | 202 | blk.22.attn_k.weight | 0x96699720 | 0x1b8000 |
306
+ | 203 | blk.22.attn_norm.weight | 0x96851720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0x96855720 | 0x900000 |
308
+ | 205 | blk.22.attn_q.weight | 0x97155720 | 0x6e0000 |
309
+ | 206 | blk.22.attn_v.weight | 0x97835720 | 0x240000 |
310
+ | 207 | blk.22.ffn_down.weight | 0x97a75720 | 0x1f80000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0x999f5720 | 0x1810000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0x9b205720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0x9b209720 | 0x1810000 |
314
+ | 211 | blk.23.attn_k.weight | 0x9ca19720 | 0x1b8000 |
315
+ | 212 | blk.23.attn_norm.weight | 0x9cbd1720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0x9cbd5720 | 0x900000 |
317
+ | 214 | blk.23.attn_q.weight | 0x9d4d5720 | 0x6e0000 |
318
+ | 215 | blk.23.attn_v.weight | 0x9dbb5720 | 0x240000 |
319
+ | 216 | blk.23.ffn_down.weight | 0x9ddf5720 | 0x1f80000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0x9fd75720 | 0x1810000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0xa1585720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0xa1589720 | 0x1810000 |
323
+ | 220 | blk.24.attn_k.weight | 0xa2d99720 | 0x1b8000 |
324
+ | 221 | blk.24.attn_norm.weight | 0xa2f51720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0xa2f55720 | 0x900000 |
326
+ | 223 | blk.24.attn_q.weight | 0xa3855720 | 0x6e0000 |
327
+ | 224 | blk.24.attn_v.weight | 0xa3f35720 | 0x240000 |
328
+ | 225 | blk.24.ffn_down.weight | 0xa4175720 | 0x1f80000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0xa60f5720 | 0x1810000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0xa7905720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0xa7909720 | 0x1810000 |
332
+ | 229 | blk.25.attn_k.weight | 0xa9119720 | 0x1b8000 |
333
+ | 230 | blk.25.attn_norm.weight | 0xa92d1720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0xa92d5720 | 0x900000 |
335
+ | 232 | blk.25.attn_q.weight | 0xa9bd5720 | 0x6e0000 |
336
+ | 233 | blk.25.attn_v.weight | 0xaa2b5720 | 0x240000 |
337
+ | 234 | blk.25.ffn_down.weight | 0xaa4f5720 | 0x1f80000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0xac475720 | 0x1810000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0xadc85720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0xadc89720 | 0x1810000 |
341
+ | 238 | blk.26.attn_k.weight | 0xaf499720 | 0x1b8000 |
342
+ | 239 | blk.26.attn_norm.weight | 0xaf651720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0xaf655720 | 0x900000 |
344
+ | 241 | blk.26.attn_q.weight | 0xaff55720 | 0x6e0000 |
345
+ | 242 | blk.26.attn_v.weight | 0xb0635720 | 0x240000 |
346
+ | 243 | blk.26.ffn_down.weight | 0xb0875720 | 0x1f80000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0xb27f5720 | 0x1810000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0xb4005720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0xb4009720 | 0x1810000 |
350
+ | 247 | blk.27.attn_k.weight | 0xb5819720 | 0x1b8000 |
351
+ | 248 | blk.27.attn_norm.weight | 0xb59d1720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0xb59d5720 | 0x900000 |
353
+ | 250 | blk.27.attn_q.weight | 0xb62d5720 | 0x6e0000 |
354
+ | 251 | blk.27.attn_v.weight | 0xb69b5720 | 0x240000 |
355
+ | 252 | blk.27.ffn_down.weight | 0xb6bf5720 | 0x1f80000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0xb8b75720 | 0x1810000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0xba385720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0xba389720 | 0x1810000 |
359
+ | 256 | blk.28.attn_k.weight | 0xbbb99720 | 0x1b8000 |
360
+ | 257 | blk.28.attn_norm.weight | 0xbbd51720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0xbbd55720 | 0x900000 |
362
+ | 259 | blk.28.attn_q.weight | 0xbc655720 | 0x6e0000 |
363
+ | 260 | blk.28.attn_v.weight | 0xbcd35720 | 0x240000 |
364
+ | 261 | blk.28.ffn_down.weight | 0xbcf75720 | 0x1f80000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0xbeef5720 | 0x1810000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0xc0705720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0xc0709720 | 0x1810000 |
368
+ | 265 | blk.29.attn_k.weight | 0xc1f19720 | 0x1b8000 |
369
+ | 266 | blk.29.attn_norm.weight | 0xc20d1720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0xc20d5720 | 0x900000 |
371
+ | 268 | blk.29.attn_q.weight | 0xc29d5720 | 0x6e0000 |
372
+ | 269 | blk.29.attn_v.weight | 0xc30b5720 | 0x240000 |
373
+ | 270 | blk.29.ffn_down.weight | 0xc32f5720 | 0x1f80000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0xc5275720 | 0x1810000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0xc6a85720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0xc6a89720 | 0x1810000 |
377
+ | 274 | blk.30.attn_k.weight | 0xc8299720 | 0x1b8000 |
378
+ | 275 | blk.30.attn_norm.weight | 0xc8451720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0xc8455720 | 0x900000 |
380
+ | 277 | blk.30.attn_q.weight | 0xc8d55720 | 0x6e0000 |
381
+ | 278 | blk.30.attn_v.weight | 0xc9435720 | 0x240000 |
382
+ | 279 | blk.30.ffn_down.weight | 0xc9675720 | 0x1f80000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0xcb5f5720 | 0x1810000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0xcce05720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0xcce09720 | 0x1810000 |
386
+ | 283 | blk.31.attn_k.weight | 0xce619720 | 0x150000 |
387
+ | 284 | blk.31.attn_norm.weight | 0xce769720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0xce76d720 | 0x900000 |
389
+ | 286 | blk.31.attn_q.weight | 0xcf06d720 | 0x540000 |
390
+ | 287 | blk.31.attn_v.weight | 0xcf5ad720 | 0x1b8000 |
391
+ | 288 | blk.31.ffn_down.weight | 0xcf765720 | 0x1f80000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0xd16e5720 | 0x1810000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0xd2ef5720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0xd2ef9720 | 0x1810000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q3_K |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q3_K |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-Q3_K_S.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-Q3_K_S.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 11 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-Q3\_K\_S.gguf - GGUF Internal File Dump](#watt-tool-8b-q3_k_sgguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0xd746000 |
104
+ | 1 | output_norm.weight | 0xdebf620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0xdec3620 | 0x100 |
106
+ | 3 | token_embd.weight | 0xdec3720 | 0xa464000 |
107
+ | 4 | blk.0.attn_k.weight | 0x18327720 | 0x150000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x18477720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x1847b720 | 0x6e0000 |
110
+ | 7 | blk.0.attn_q.weight | 0x18b5b720 | 0x540000 |
111
+ | 8 | blk.0.attn_v.weight | 0x1909b720 | 0x1b8000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x19253720 | 0x1810000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x1aa63720 | 0x1260000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x1bcc3720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x1bcc7720 | 0x1260000 |
116
+ | 13 | blk.1.attn_k.weight | 0x1cf27720 | 0x150000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x1d077720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x1d07b720 | 0x6e0000 |
119
+ | 16 | blk.1.attn_q.weight | 0x1d75b720 | 0x540000 |
120
+ | 17 | blk.1.attn_v.weight | 0x1dc9b720 | 0x1b8000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x1de53720 | 0x1f80000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x1fdd3720 | 0x1260000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x21033720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x21037720 | 0x1260000 |
125
+ | 22 | blk.2.attn_k.weight | 0x22297720 | 0x150000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x223e7720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x223eb720 | 0x6e0000 |
128
+ | 25 | blk.2.attn_q.weight | 0x22acb720 | 0x540000 |
129
+ | 26 | blk.2.attn_v.weight | 0x2300b720 | 0x1b8000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x231c3720 | 0x1810000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x249d3720 | 0x1260000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x25c33720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x25c37720 | 0x1260000 |
134
+ | 31 | blk.3.attn_k.weight | 0x26e97720 | 0x150000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x26fe7720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x26feb720 | 0x6e0000 |
137
+ | 34 | blk.3.attn_q.weight | 0x276cb720 | 0x540000 |
138
+ | 35 | blk.3.attn_v.weight | 0x27c0b720 | 0x1b8000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x27dc3720 | 0x1810000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x295d3720 | 0x1260000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x2a833720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x2a837720 | 0x1260000 |
143
+ | 40 | blk.4.attn_k.weight | 0x2ba97720 | 0x150000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x2bbe7720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x2bbeb720 | 0x6e0000 |
146
+ | 43 | blk.4.attn_q.weight | 0x2c2cb720 | 0x540000 |
147
+ | 44 | blk.4.attn_v.weight | 0x2c80b720 | 0x1b8000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x2c9c3720 | 0x1810000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x2e1d3720 | 0x1260000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x2f433720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x2f437720 | 0x1260000 |
152
+ | 49 | blk.5.attn_k.weight | 0x30697720 | 0x150000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x307e7720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x307eb720 | 0x6e0000 |
155
+ | 52 | blk.5.attn_q.weight | 0x30ecb720 | 0x540000 |
156
+ | 53 | blk.5.attn_v.weight | 0x3140b720 | 0x1b8000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x315c3720 | 0x1810000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x32dd3720 | 0x1260000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x34033720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x34037720 | 0x1260000 |
161
+ | 58 | blk.6.attn_k.weight | 0x35297720 | 0x150000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x353e7720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x353eb720 | 0x6e0000 |
164
+ | 61 | blk.6.attn_q.weight | 0x35acb720 | 0x540000 |
165
+ | 62 | blk.6.attn_v.weight | 0x3600b720 | 0x1b8000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x361c3720 | 0x1810000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x379d3720 | 0x1260000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x38c33720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x38c37720 | 0x1260000 |
170
+ | 67 | blk.7.attn_k.weight | 0x39e97720 | 0x150000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x39fe7720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x39feb720 | 0x6e0000 |
173
+ | 70 | blk.7.attn_q.weight | 0x3a6cb720 | 0x540000 |
174
+ | 71 | blk.7.attn_v.weight | 0x3ac0b720 | 0x1b8000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x3adc3720 | 0x1810000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x3c5d3720 | 0x1260000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x3d833720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x3d837720 | 0x1260000 |
179
+ | 76 | blk.8.attn_k.weight | 0x3ea97720 | 0x150000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x3ebe7720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x3ebeb720 | 0x6e0000 |
182
+ | 79 | blk.8.attn_q.weight | 0x3f2cb720 | 0x540000 |
183
+ | 80 | blk.8.attn_v.weight | 0x3f80b720 | 0x1b8000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x3f9c3720 | 0x1810000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x411d3720 | 0x1260000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x42433720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x42437720 | 0x1260000 |
188
+ | 85 | blk.9.attn_k.weight | 0x43697720 | 0x150000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x437e7720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x437eb720 | 0x6e0000 |
191
+ | 88 | blk.9.attn_q.weight | 0x43ecb720 | 0x540000 |
192
+ | 89 | blk.9.attn_v.weight | 0x4440b720 | 0x1b8000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x445c3720 | 0x1810000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x45dd3720 | 0x1260000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0x47033720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0x47037720 | 0x1260000 |
197
+ | 94 | blk.10.attn_k.weight | 0x48297720 | 0x150000 |
198
+ | 95 | blk.10.attn_norm.weight | 0x483e7720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0x483eb720 | 0x6e0000 |
200
+ | 97 | blk.10.attn_q.weight | 0x48acb720 | 0x540000 |
201
+ | 98 | blk.10.attn_v.weight | 0x4900b720 | 0x1b8000 |
202
+ | 99 | blk.10.ffn_down.weight | 0x491c3720 | 0x1810000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0x4a9d3720 | 0x1260000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0x4bc33720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0x4bc37720 | 0x1260000 |
206
+ | 103 | blk.11.attn_k.weight | 0x4ce97720 | 0x150000 |
207
+ | 104 | blk.11.attn_norm.weight | 0x4cfe7720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0x4cfeb720 | 0x6e0000 |
209
+ | 106 | blk.11.attn_q.weight | 0x4d6cb720 | 0x540000 |
210
+ | 107 | blk.11.attn_v.weight | 0x4dc0b720 | 0x1b8000 |
211
+ | 108 | blk.11.ffn_down.weight | 0x4ddc3720 | 0x1810000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0x4f5d3720 | 0x1260000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0x50833720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0x50837720 | 0x1260000 |
215
+ | 112 | blk.12.attn_k.weight | 0x51a97720 | 0x150000 |
216
+ | 113 | blk.12.attn_norm.weight | 0x51be7720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0x51beb720 | 0x6e0000 |
218
+ | 115 | blk.12.attn_q.weight | 0x522cb720 | 0x540000 |
219
+ | 116 | blk.12.attn_v.weight | 0x5280b720 | 0x1b8000 |
220
+ | 117 | blk.12.ffn_down.weight | 0x529c3720 | 0x1810000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0x541d3720 | 0x1260000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0x55433720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0x55437720 | 0x1260000 |
224
+ | 121 | blk.13.attn_k.weight | 0x56697720 | 0x1b8000 |
225
+ | 122 | blk.13.attn_norm.weight | 0x5684f720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0x56853720 | 0x6e0000 |
227
+ | 124 | blk.13.attn_q.weight | 0x56f33720 | 0x6e0000 |
228
+ | 125 | blk.13.attn_v.weight | 0x57613720 | 0x1b8000 |
229
+ | 126 | blk.13.ffn_down.weight | 0x577cb720 | 0x1810000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0x58fdb720 | 0x1260000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0x5a23b720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0x5a23f720 | 0x1260000 |
233
+ | 130 | blk.14.attn_k.weight | 0x5b49f720 | 0x1b8000 |
234
+ | 131 | blk.14.attn_norm.weight | 0x5b657720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0x5b65b720 | 0x6e0000 |
236
+ | 133 | blk.14.attn_q.weight | 0x5bd3b720 | 0x6e0000 |
237
+ | 134 | blk.14.attn_v.weight | 0x5c41b720 | 0x1b8000 |
238
+ | 135 | blk.14.ffn_down.weight | 0x5c5d3720 | 0x1810000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0x5dde3720 | 0x1260000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0x5f043720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0x5f047720 | 0x1260000 |
242
+ | 139 | blk.15.attn_k.weight | 0x602a7720 | 0x150000 |
243
+ | 140 | blk.15.attn_norm.weight | 0x603f7720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0x603fb720 | 0x6e0000 |
245
+ | 142 | blk.15.attn_q.weight | 0x60adb720 | 0x540000 |
246
+ | 143 | blk.15.attn_v.weight | 0x6101b720 | 0x1b8000 |
247
+ | 144 | blk.15.ffn_down.weight | 0x611d3720 | 0x1810000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0x629e3720 | 0x1260000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0x63c43720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0x63c47720 | 0x1260000 |
251
+ | 148 | blk.16.attn_k.weight | 0x64ea7720 | 0x1b8000 |
252
+ | 149 | blk.16.attn_norm.weight | 0x6505f720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0x65063720 | 0x6e0000 |
254
+ | 151 | blk.16.attn_q.weight | 0x65743720 | 0x6e0000 |
255
+ | 152 | blk.16.attn_v.weight | 0x65e23720 | 0x1b8000 |
256
+ | 153 | blk.16.ffn_down.weight | 0x65fdb720 | 0x1810000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0x677eb720 | 0x1810000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0x68ffb720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0x68fff720 | 0x1810000 |
260
+ | 157 | blk.17.attn_k.weight | 0x6a80f720 | 0x150000 |
261
+ | 158 | blk.17.attn_norm.weight | 0x6a95f720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0x6a963720 | 0x6e0000 |
263
+ | 160 | blk.17.attn_q.weight | 0x6b043720 | 0x540000 |
264
+ | 161 | blk.17.attn_v.weight | 0x6b583720 | 0x1b8000 |
265
+ | 162 | blk.17.ffn_down.weight | 0x6b73b720 | 0x1f80000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0x6d6bb720 | 0x1810000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0x6eecb720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0x6eecf720 | 0x1810000 |
269
+ | 166 | blk.18.attn_k.weight | 0x706df720 | 0x1b8000 |
270
+ | 167 | blk.18.attn_norm.weight | 0x70897720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0x7089b720 | 0x6e0000 |
272
+ | 169 | blk.18.attn_q.weight | 0x70f7b720 | 0x6e0000 |
273
+ | 170 | blk.18.attn_v.weight | 0x7165b720 | 0x1b8000 |
274
+ | 171 | blk.18.ffn_down.weight | 0x71813720 | 0x1f80000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0x73793720 | 0x1810000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0x74fa3720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0x74fa7720 | 0x1810000 |
278
+ | 175 | blk.19.attn_k.weight | 0x767b7720 | 0x1b8000 |
279
+ | 176 | blk.19.attn_norm.weight | 0x7696f720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0x76973720 | 0x6e0000 |
281
+ | 178 | blk.19.attn_q.weight | 0x77053720 | 0x6e0000 |
282
+ | 179 | blk.19.attn_v.weight | 0x77733720 | 0x1b8000 |
283
+ | 180 | blk.19.ffn_down.weight | 0x778eb720 | 0x1f80000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0x7986b720 | 0x1810000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0x7b07b720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0x7b07f720 | 0x1810000 |
287
+ | 184 | blk.20.attn_k.weight | 0x7c88f720 | 0x1b8000 |
288
+ | 185 | blk.20.attn_norm.weight | 0x7ca47720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0x7ca4b720 | 0x6e0000 |
290
+ | 187 | blk.20.attn_q.weight | 0x7d12b720 | 0x6e0000 |
291
+ | 188 | blk.20.attn_v.weight | 0x7d80b720 | 0x1b8000 |
292
+ | 189 | blk.20.ffn_down.weight | 0x7d9c3720 | 0x1f80000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0x7f943720 | 0x1810000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0x81153720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0x81157720 | 0x1810000 |
296
+ | 193 | blk.21.attn_k.weight | 0x82967720 | 0x1b8000 |
297
+ | 194 | blk.21.attn_norm.weight | 0x82b1f720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0x82b23720 | 0x6e0000 |
299
+ | 196 | blk.21.attn_q.weight | 0x83203720 | 0x6e0000 |
300
+ | 197 | blk.21.attn_v.weight | 0x838e3720 | 0x1b8000 |
301
+ | 198 | blk.21.ffn_down.weight | 0x83a9b720 | 0x1f80000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0x85a1b720 | 0x1810000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0x8722b720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0x8722f720 | 0x1810000 |
305
+ | 202 | blk.22.attn_k.weight | 0x88a3f720 | 0x1b8000 |
306
+ | 203 | blk.22.attn_norm.weight | 0x88bf7720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0x88bfb720 | 0x6e0000 |
308
+ | 205 | blk.22.attn_q.weight | 0x892db720 | 0x6e0000 |
309
+ | 206 | blk.22.attn_v.weight | 0x899bb720 | 0x1b8000 |
310
+ | 207 | blk.22.ffn_down.weight | 0x89b73720 | 0x1f80000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0x8baf3720 | 0x1810000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0x8d303720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0x8d307720 | 0x1810000 |
314
+ | 211 | blk.23.attn_k.weight | 0x8eb17720 | 0x1b8000 |
315
+ | 212 | blk.23.attn_norm.weight | 0x8eccf720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0x8ecd3720 | 0x6e0000 |
317
+ | 214 | blk.23.attn_q.weight | 0x8f3b3720 | 0x6e0000 |
318
+ | 215 | blk.23.attn_v.weight | 0x8fa93720 | 0x1b8000 |
319
+ | 216 | blk.23.ffn_down.weight | 0x8fc4b720 | 0x1f80000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0x91bcb720 | 0x1810000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0x933db720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0x933df720 | 0x1810000 |
323
+ | 220 | blk.24.attn_k.weight | 0x94bef720 | 0x1b8000 |
324
+ | 221 | blk.24.attn_norm.weight | 0x94da7720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0x94dab720 | 0x6e0000 |
326
+ | 223 | blk.24.attn_q.weight | 0x9548b720 | 0x6e0000 |
327
+ | 224 | blk.24.attn_v.weight | 0x95b6b720 | 0x1b8000 |
328
+ | 225 | blk.24.ffn_down.weight | 0x95d23720 | 0x1f80000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0x97ca3720 | 0x1810000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0x994b3720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0x994b7720 | 0x1810000 |
332
+ | 229 | blk.25.attn_k.weight | 0x9acc7720 | 0x1b8000 |
333
+ | 230 | blk.25.attn_norm.weight | 0x9ae7f720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0x9ae83720 | 0x6e0000 |
335
+ | 232 | blk.25.attn_q.weight | 0x9b563720 | 0x6e0000 |
336
+ | 233 | blk.25.attn_v.weight | 0x9bc43720 | 0x1b8000 |
337
+ | 234 | blk.25.ffn_down.weight | 0x9bdfb720 | 0x1f80000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0x9dd7b720 | 0x1810000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0x9f58b720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0x9f58f720 | 0x1810000 |
341
+ | 238 | blk.26.attn_k.weight | 0xa0d9f720 | 0x1b8000 |
342
+ | 239 | blk.26.attn_norm.weight | 0xa0f57720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0xa0f5b720 | 0x6e0000 |
344
+ | 241 | blk.26.attn_q.weight | 0xa163b720 | 0x6e0000 |
345
+ | 242 | blk.26.attn_v.weight | 0xa1d1b720 | 0x1b8000 |
346
+ | 243 | blk.26.ffn_down.weight | 0xa1ed3720 | 0x1f80000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0xa3e53720 | 0x1810000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0xa5663720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0xa5667720 | 0x1810000 |
350
+ | 247 | blk.27.attn_k.weight | 0xa6e77720 | 0x1b8000 |
351
+ | 248 | blk.27.attn_norm.weight | 0xa702f720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0xa7033720 | 0x6e0000 |
353
+ | 250 | blk.27.attn_q.weight | 0xa7713720 | 0x6e0000 |
354
+ | 251 | blk.27.attn_v.weight | 0xa7df3720 | 0x1b8000 |
355
+ | 252 | blk.27.ffn_down.weight | 0xa7fab720 | 0x1f80000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0xa9f2b720 | 0x1810000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0xab73b720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0xab73f720 | 0x1810000 |
359
+ | 256 | blk.28.attn_k.weight | 0xacf4f720 | 0x1b8000 |
360
+ | 257 | blk.28.attn_norm.weight | 0xad107720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0xad10b720 | 0x6e0000 |
362
+ | 259 | blk.28.attn_q.weight | 0xad7eb720 | 0x6e0000 |
363
+ | 260 | blk.28.attn_v.weight | 0xadecb720 | 0x1b8000 |
364
+ | 261 | blk.28.ffn_down.weight | 0xae083720 | 0x1f80000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0xb0003720 | 0x1810000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0xb1813720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0xb1817720 | 0x1810000 |
368
+ | 265 | blk.29.attn_k.weight | 0xb3027720 | 0x1b8000 |
369
+ | 266 | blk.29.attn_norm.weight | 0xb31df720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0xb31e3720 | 0x6e0000 |
371
+ | 268 | blk.29.attn_q.weight | 0xb38c3720 | 0x6e0000 |
372
+ | 269 | blk.29.attn_v.weight | 0xb3fa3720 | 0x1b8000 |
373
+ | 270 | blk.29.ffn_down.weight | 0xb415b720 | 0x1f80000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0xb60db720 | 0x1810000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0xb78eb720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0xb78ef720 | 0x1810000 |
377
+ | 274 | blk.30.attn_k.weight | 0xb90ff720 | 0x1b8000 |
378
+ | 275 | blk.30.attn_norm.weight | 0xb92b7720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0xb92bb720 | 0x6e0000 |
380
+ | 277 | blk.30.attn_q.weight | 0xb999b720 | 0x6e0000 |
381
+ | 278 | blk.30.attn_v.weight | 0xba07b720 | 0x1b8000 |
382
+ | 279 | blk.30.ffn_down.weight | 0xba233720 | 0x1f80000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0xbc1b3720 | 0x1810000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0xbd9c3720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0xbd9c7720 | 0x1810000 |
386
+ | 283 | blk.31.attn_k.weight | 0xbf1d7720 | 0x150000 |
387
+ | 284 | blk.31.attn_norm.weight | 0xbf327720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0xbf32b720 | 0x6e0000 |
389
+ | 286 | blk.31.attn_q.weight | 0xbfa0b720 | 0x540000 |
390
+ | 287 | blk.31.attn_v.weight | 0xbff4b720 | 0x1b8000 |
391
+ | 288 | blk.31.ffn_down.weight | 0xc0103720 | 0x1f80000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0xc2083720 | 0x1810000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0xc3893720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0xc3897720 | 0x1810000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q3_K |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q2_K |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q2_K |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q3_K |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q2_K |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q2_K |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-Q4_K_M.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-Q4_K_M.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 15 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-Q4\_K\_M.gguf - GGUF Internal File Dump](#watt-tool-8b-q4_k_mgguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0x119d0000 |
104
+ | 1 | output_norm.weight | 0x12149620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0x1214d620 | 0x100 |
106
+ | 3 | token_embd.weight | 0x1214d720 | 0xd746000 |
107
+ | 4 | blk.0.attn_k.weight | 0x1f893720 | 0x1b8000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x1fa4b720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x1fa4f720 | 0x900000 |
110
+ | 7 | blk.0.attn_q.weight | 0x2034f720 | 0x6e0000 |
111
+ | 8 | blk.0.attn_v.weight | 0x20a2f720 | 0x240000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x20c6f720 | 0x2680000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x232ef720 | 0x1810000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x24aff720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x24b03720 | 0x1810000 |
116
+ | 13 | blk.1.attn_k.weight | 0x26313720 | 0x1b8000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x264cb720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x264cf720 | 0x900000 |
119
+ | 16 | blk.1.attn_q.weight | 0x26dcf720 | 0x6e0000 |
120
+ | 17 | blk.1.attn_v.weight | 0x274af720 | 0x240000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x276ef720 | 0x2680000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x29d6f720 | 0x1810000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x2b57f720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x2b583720 | 0x1810000 |
125
+ | 22 | blk.2.attn_k.weight | 0x2cd93720 | 0x1b8000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x2cf4b720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x2cf4f720 | 0x900000 |
128
+ | 25 | blk.2.attn_q.weight | 0x2d84f720 | 0x6e0000 |
129
+ | 26 | blk.2.attn_v.weight | 0x2df2f720 | 0x240000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x2e16f720 | 0x2680000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x307ef720 | 0x1810000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x31fff720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x32003720 | 0x1810000 |
134
+ | 31 | blk.3.attn_k.weight | 0x33813720 | 0x1b8000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x339cb720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x339cf720 | 0x900000 |
137
+ | 34 | blk.3.attn_q.weight | 0x342cf720 | 0x6e0000 |
138
+ | 35 | blk.3.attn_v.weight | 0x349af720 | 0x240000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x34bef720 | 0x2680000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x3726f720 | 0x1810000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x38a7f720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x38a83720 | 0x1810000 |
143
+ | 40 | blk.4.attn_k.weight | 0x3a293720 | 0x1b8000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x3a44b720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x3a44f720 | 0x900000 |
146
+ | 43 | blk.4.attn_q.weight | 0x3ad4f720 | 0x6e0000 |
147
+ | 44 | blk.4.attn_v.weight | 0x3b42f720 | 0x240000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x3b66f720 | 0x2680000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x3dcef720 | 0x1810000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x3f4ff720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x3f503720 | 0x1810000 |
152
+ | 49 | blk.5.attn_k.weight | 0x40d13720 | 0x1b8000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x40ecb720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x40ecf720 | 0x900000 |
155
+ | 52 | blk.5.attn_q.weight | 0x417cf720 | 0x6e0000 |
156
+ | 53 | blk.5.attn_v.weight | 0x41eaf720 | 0x240000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x420ef720 | 0x2680000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x4476f720 | 0x1810000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x45f7f720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x45f83720 | 0x1810000 |
161
+ | 58 | blk.6.attn_k.weight | 0x47793720 | 0x1b8000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x4794b720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x4794f720 | 0x900000 |
164
+ | 61 | blk.6.attn_q.weight | 0x4824f720 | 0x6e0000 |
165
+ | 62 | blk.6.attn_v.weight | 0x4892f720 | 0x240000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x48b6f720 | 0x2680000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x4b1ef720 | 0x1810000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x4c9ff720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x4ca03720 | 0x1810000 |
170
+ | 67 | blk.7.attn_k.weight | 0x4e213720 | 0x1b8000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x4e3cb720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x4e3cf720 | 0x900000 |
173
+ | 70 | blk.7.attn_q.weight | 0x4eccf720 | 0x6e0000 |
174
+ | 71 | blk.7.attn_v.weight | 0x4f3af720 | 0x240000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x4f5ef720 | 0x2680000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x51c6f720 | 0x1810000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x5347f720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x53483720 | 0x1810000 |
179
+ | 76 | blk.8.attn_k.weight | 0x54c93720 | 0x1b8000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x54e4b720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x54e4f720 | 0x900000 |
182
+ | 79 | blk.8.attn_q.weight | 0x5574f720 | 0x6e0000 |
183
+ | 80 | blk.8.attn_v.weight | 0x55e2f720 | 0x240000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x5606f720 | 0x2680000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x586ef720 | 0x1810000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x59eff720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x59f03720 | 0x1810000 |
188
+ | 85 | blk.9.attn_k.weight | 0x5b713720 | 0x1b8000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x5b8cb720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x5b8cf720 | 0x900000 |
191
+ | 88 | blk.9.attn_q.weight | 0x5c1cf720 | 0x6e0000 |
192
+ | 89 | blk.9.attn_v.weight | 0x5c8af720 | 0x240000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x5caef720 | 0x2680000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x5f16f720 | 0x1810000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0x6097f720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0x60983720 | 0x1810000 |
197
+ | 94 | blk.10.attn_k.weight | 0x62193720 | 0x1b8000 |
198
+ | 95 | blk.10.attn_norm.weight | 0x6234b720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0x6234f720 | 0x900000 |
200
+ | 97 | blk.10.attn_q.weight | 0x62c4f720 | 0x6e0000 |
201
+ | 98 | blk.10.attn_v.weight | 0x6332f720 | 0x240000 |
202
+ | 99 | blk.10.ffn_down.weight | 0x6356f720 | 0x2680000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0x65bef720 | 0x1810000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0x673ff720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0x67403720 | 0x1810000 |
206
+ | 103 | blk.11.attn_k.weight | 0x68c13720 | 0x1b8000 |
207
+ | 104 | blk.11.attn_norm.weight | 0x68dcb720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0x68dcf720 | 0x900000 |
209
+ | 106 | blk.11.attn_q.weight | 0x696cf720 | 0x6e0000 |
210
+ | 107 | blk.11.attn_v.weight | 0x69daf720 | 0x240000 |
211
+ | 108 | blk.11.ffn_down.weight | 0x69fef720 | 0x2680000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0x6c66f720 | 0x1810000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0x6de7f720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0x6de83720 | 0x1810000 |
215
+ | 112 | blk.12.attn_k.weight | 0x6f693720 | 0x1b8000 |
216
+ | 113 | blk.12.attn_norm.weight | 0x6f84b720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0x6f84f720 | 0x900000 |
218
+ | 115 | blk.12.attn_q.weight | 0x7014f720 | 0x6e0000 |
219
+ | 116 | blk.12.attn_v.weight | 0x7082f720 | 0x240000 |
220
+ | 117 | blk.12.ffn_down.weight | 0x70a6f720 | 0x2680000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0x730ef720 | 0x1810000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0x748ff720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0x74903720 | 0x1810000 |
224
+ | 121 | blk.13.attn_k.weight | 0x76113720 | 0x240000 |
225
+ | 122 | blk.13.attn_norm.weight | 0x76353720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0x76357720 | 0x900000 |
227
+ | 124 | blk.13.attn_q.weight | 0x76c57720 | 0x900000 |
228
+ | 125 | blk.13.attn_v.weight | 0x77557720 | 0x2c0000 |
229
+ | 126 | blk.13.ffn_down.weight | 0x77817720 | 0x2680000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0x79e97720 | 0x1810000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0x7b6a7720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0x7b6ab720 | 0x1810000 |
233
+ | 130 | blk.14.attn_k.weight | 0x7cebb720 | 0x240000 |
234
+ | 131 | blk.14.attn_norm.weight | 0x7d0fb720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0x7d0ff720 | 0x900000 |
236
+ | 133 | blk.14.attn_q.weight | 0x7d9ff720 | 0x900000 |
237
+ | 134 | blk.14.attn_v.weight | 0x7e2ff720 | 0x2c0000 |
238
+ | 135 | blk.14.ffn_down.weight | 0x7e5bf720 | 0x2680000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0x80c3f720 | 0x1810000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0x8244f720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0x82453720 | 0x1810000 |
242
+ | 139 | blk.15.attn_k.weight | 0x83c63720 | 0x1b8000 |
243
+ | 140 | blk.15.attn_norm.weight | 0x83e1b720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0x83e1f720 | 0x900000 |
245
+ | 142 | blk.15.attn_q.weight | 0x8471f720 | 0x6e0000 |
246
+ | 143 | blk.15.attn_v.weight | 0x84dff720 | 0x240000 |
247
+ | 144 | blk.15.ffn_down.weight | 0x8503f720 | 0x2680000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0x876bf720 | 0x1810000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0x88ecf720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0x88ed3720 | 0x1810000 |
251
+ | 148 | blk.16.attn_k.weight | 0x8a6e3720 | 0x240000 |
252
+ | 149 | blk.16.attn_norm.weight | 0x8a923720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0x8a927720 | 0x900000 |
254
+ | 151 | blk.16.attn_q.weight | 0x8b227720 | 0x900000 |
255
+ | 152 | blk.16.attn_v.weight | 0x8bb27720 | 0x2c0000 |
256
+ | 153 | blk.16.ffn_down.weight | 0x8bde7720 | 0x2680000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0x8e467720 | 0x1f80000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0x903e7720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0x903eb720 | 0x1f80000 |
260
+ | 157 | blk.17.attn_k.weight | 0x9236b720 | 0x1b8000 |
261
+ | 158 | blk.17.attn_norm.weight | 0x92523720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0x92527720 | 0x900000 |
263
+ | 160 | blk.17.attn_q.weight | 0x92e27720 | 0x6e0000 |
264
+ | 161 | blk.17.attn_v.weight | 0x93507720 | 0x240000 |
265
+ | 162 | blk.17.ffn_down.weight | 0x93747720 | 0x2680000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0x95dc7720 | 0x1f80000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0x97d47720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0x97d4b720 | 0x1f80000 |
269
+ | 166 | blk.18.attn_k.weight | 0x99ccb720 | 0x240000 |
270
+ | 167 | blk.18.attn_norm.weight | 0x99f0b720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0x99f0f720 | 0x900000 |
272
+ | 169 | blk.18.attn_q.weight | 0x9a80f720 | 0x900000 |
273
+ | 170 | blk.18.attn_v.weight | 0x9b10f720 | 0x2c0000 |
274
+ | 171 | blk.18.ffn_down.weight | 0x9b3cf720 | 0x2680000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0x9da4f720 | 0x1f80000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0x9f9cf720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0x9f9d3720 | 0x1f80000 |
278
+ | 175 | blk.19.attn_k.weight | 0xa1953720 | 0x240000 |
279
+ | 176 | blk.19.attn_norm.weight | 0xa1b93720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0xa1b97720 | 0x900000 |
281
+ | 178 | blk.19.attn_q.weight | 0xa2497720 | 0x900000 |
282
+ | 179 | blk.19.attn_v.weight | 0xa2d97720 | 0x2c0000 |
283
+ | 180 | blk.19.ffn_down.weight | 0xa3057720 | 0x2680000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0xa56d7720 | 0x1f80000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0xa7657720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0xa765b720 | 0x1f80000 |
287
+ | 184 | blk.20.attn_k.weight | 0xa95db720 | 0x240000 |
288
+ | 185 | blk.20.attn_norm.weight | 0xa981b720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0xa981f720 | 0x900000 |
290
+ | 187 | blk.20.attn_q.weight | 0xaa11f720 | 0x900000 |
291
+ | 188 | blk.20.attn_v.weight | 0xaaa1f720 | 0x2c0000 |
292
+ | 189 | blk.20.ffn_down.weight | 0xaacdf720 | 0x2680000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0xad35f720 | 0x1f80000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0xaf2df720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0xaf2e3720 | 0x1f80000 |
296
+ | 193 | blk.21.attn_k.weight | 0xb1263720 | 0x240000 |
297
+ | 194 | blk.21.attn_norm.weight | 0xb14a3720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0xb14a7720 | 0x900000 |
299
+ | 196 | blk.21.attn_q.weight | 0xb1da7720 | 0x900000 |
300
+ | 197 | blk.21.attn_v.weight | 0xb26a7720 | 0x2c0000 |
301
+ | 198 | blk.21.ffn_down.weight | 0xb2967720 | 0x2680000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0xb4fe7720 | 0x1f80000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0xb6f67720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0xb6f6b720 | 0x1f80000 |
305
+ | 202 | blk.22.attn_k.weight | 0xb8eeb720 | 0x240000 |
306
+ | 203 | blk.22.attn_norm.weight | 0xb912b720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0xb912f720 | 0x900000 |
308
+ | 205 | blk.22.attn_q.weight | 0xb9a2f720 | 0x900000 |
309
+ | 206 | blk.22.attn_v.weight | 0xba32f720 | 0x2c0000 |
310
+ | 207 | blk.22.ffn_down.weight | 0xba5ef720 | 0x2680000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0xbcc6f720 | 0x1f80000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0xbebef720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0xbebf3720 | 0x1f80000 |
314
+ | 211 | blk.23.attn_k.weight | 0xc0b73720 | 0x240000 |
315
+ | 212 | blk.23.attn_norm.weight | 0xc0db3720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0xc0db7720 | 0x900000 |
317
+ | 214 | blk.23.attn_q.weight | 0xc16b7720 | 0x900000 |
318
+ | 215 | blk.23.attn_v.weight | 0xc1fb7720 | 0x2c0000 |
319
+ | 216 | blk.23.ffn_down.weight | 0xc2277720 | 0x2680000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0xc48f7720 | 0x1f80000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0xc6877720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0xc687b720 | 0x1f80000 |
323
+ | 220 | blk.24.attn_k.weight | 0xc87fb720 | 0x240000 |
324
+ | 221 | blk.24.attn_norm.weight | 0xc8a3b720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0xc8a3f720 | 0x900000 |
326
+ | 223 | blk.24.attn_q.weight | 0xc933f720 | 0x900000 |
327
+ | 224 | blk.24.attn_v.weight | 0xc9c3f720 | 0x2c0000 |
328
+ | 225 | blk.24.ffn_down.weight | 0xc9eff720 | 0x2680000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0xcc57f720 | 0x1f80000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0xce4ff720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0xce503720 | 0x1f80000 |
332
+ | 229 | blk.25.attn_k.weight | 0xd0483720 | 0x240000 |
333
+ | 230 | blk.25.attn_norm.weight | 0xd06c3720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0xd06c7720 | 0x900000 |
335
+ | 232 | blk.25.attn_q.weight | 0xd0fc7720 | 0x900000 |
336
+ | 233 | blk.25.attn_v.weight | 0xd18c7720 | 0x2c0000 |
337
+ | 234 | blk.25.ffn_down.weight | 0xd1b87720 | 0x2680000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0xd4207720 | 0x1f80000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0xd6187720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0xd618b720 | 0x1f80000 |
341
+ | 238 | blk.26.attn_k.weight | 0xd810b720 | 0x240000 |
342
+ | 239 | blk.26.attn_norm.weight | 0xd834b720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0xd834f720 | 0x900000 |
344
+ | 241 | blk.26.attn_q.weight | 0xd8c4f720 | 0x900000 |
345
+ | 242 | blk.26.attn_v.weight | 0xd954f720 | 0x2c0000 |
346
+ | 243 | blk.26.ffn_down.weight | 0xd980f720 | 0x2680000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0xdbe8f720 | 0x1f80000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0xdde0f720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0xdde13720 | 0x1f80000 |
350
+ | 247 | blk.27.attn_k.weight | 0xdfd93720 | 0x240000 |
351
+ | 248 | blk.27.attn_norm.weight | 0xdffd3720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0xdffd7720 | 0x900000 |
353
+ | 250 | blk.27.attn_q.weight | 0xe08d7720 | 0x900000 |
354
+ | 251 | blk.27.attn_v.weight | 0xe11d7720 | 0x2c0000 |
355
+ | 252 | blk.27.ffn_down.weight | 0xe1497720 | 0x2680000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0xe3b17720 | 0x1f80000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0xe5a97720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0xe5a9b720 | 0x1f80000 |
359
+ | 256 | blk.28.attn_k.weight | 0xe7a1b720 | 0x240000 |
360
+ | 257 | blk.28.attn_norm.weight | 0xe7c5b720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0xe7c5f720 | 0x900000 |
362
+ | 259 | blk.28.attn_q.weight | 0xe855f720 | 0x900000 |
363
+ | 260 | blk.28.attn_v.weight | 0xe8e5f720 | 0x2c0000 |
364
+ | 261 | blk.28.ffn_down.weight | 0xe911f720 | 0x2680000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0xeb79f720 | 0x1f80000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0xed71f720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0xed723720 | 0x1f80000 |
368
+ | 265 | blk.29.attn_k.weight | 0xef6a3720 | 0x240000 |
369
+ | 266 | blk.29.attn_norm.weight | 0xef8e3720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0xef8e7720 | 0x900000 |
371
+ | 268 | blk.29.attn_q.weight | 0xf01e7720 | 0x900000 |
372
+ | 269 | blk.29.attn_v.weight | 0xf0ae7720 | 0x2c0000 |
373
+ | 270 | blk.29.ffn_down.weight | 0xf0da7720 | 0x2680000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0xf3427720 | 0x1f80000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0xf53a7720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0xf53ab720 | 0x1f80000 |
377
+ | 274 | blk.30.attn_k.weight | 0xf732b720 | 0x240000 |
378
+ | 275 | blk.30.attn_norm.weight | 0xf756b720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0xf756f720 | 0x900000 |
380
+ | 277 | blk.30.attn_q.weight | 0xf7e6f720 | 0x900000 |
381
+ | 278 | blk.30.attn_v.weight | 0xf876f720 | 0x2c0000 |
382
+ | 279 | blk.30.ffn_down.weight | 0xf8a2f720 | 0x2680000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0xfb0af720 | 0x1f80000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0xfd02f720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0xfd033720 | 0x1f80000 |
386
+ | 283 | blk.31.attn_k.weight | 0xfefb3720 | 0x1b8000 |
387
+ | 284 | blk.31.attn_norm.weight | 0xff16b720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0xff16f720 | 0x900000 |
389
+ | 286 | blk.31.attn_q.weight | 0xffa6f720 | 0x6e0000 |
390
+ | 287 | blk.31.attn_v.weight | 0x10014f720 | 0x240000 |
391
+ | 288 | blk.31.ffn_down.weight | 0x10038f720 | 0x2680000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0x102a0f720 | 0x1f80000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0x10498f720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0x104993720 | 0x1f80000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q4_K |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q3_K |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-Q4_K_S.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-Q4_K_S.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 14 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-Q4\_K\_S.gguf - GGUF Internal File Dump](#watt-tool-8b-q4_k_sgguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0x119d0000 |
104
+ | 1 | output_norm.weight | 0x12149620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0x1214d620 | 0x100 |
106
+ | 3 | token_embd.weight | 0x1214d720 | 0xd746000 |
107
+ | 4 | blk.0.attn_k.weight | 0x1f893720 | 0x1b8000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x1fa4b720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x1fa4f720 | 0x900000 |
110
+ | 7 | blk.0.attn_q.weight | 0x2034f720 | 0x6e0000 |
111
+ | 8 | blk.0.attn_v.weight | 0x20a2f720 | 0x240000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x20c6f720 | 0x1f80000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x22bef720 | 0x1810000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x243ff720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x24403720 | 0x1810000 |
116
+ | 13 | blk.1.attn_k.weight | 0x25c13720 | 0x1b8000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x25dcb720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x25dcf720 | 0x900000 |
119
+ | 16 | blk.1.attn_q.weight | 0x266cf720 | 0x6e0000 |
120
+ | 17 | blk.1.attn_v.weight | 0x26daf720 | 0x240000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x26fef720 | 0x2680000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x2966f720 | 0x1810000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x2ae7f720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x2ae83720 | 0x1810000 |
125
+ | 22 | blk.2.attn_k.weight | 0x2c693720 | 0x1b8000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x2c84b720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x2c84f720 | 0x900000 |
128
+ | 25 | blk.2.attn_q.weight | 0x2d14f720 | 0x6e0000 |
129
+ | 26 | blk.2.attn_v.weight | 0x2d82f720 | 0x240000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x2da6f720 | 0x1f80000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x2f9ef720 | 0x1810000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x311ff720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x31203720 | 0x1810000 |
134
+ | 31 | blk.3.attn_k.weight | 0x32a13720 | 0x1b8000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x32bcb720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x32bcf720 | 0x900000 |
137
+ | 34 | blk.3.attn_q.weight | 0x334cf720 | 0x6e0000 |
138
+ | 35 | blk.3.attn_v.weight | 0x33baf720 | 0x240000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x33def720 | 0x1f80000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x35d6f720 | 0x1810000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x3757f720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x37583720 | 0x1810000 |
143
+ | 40 | blk.4.attn_k.weight | 0x38d93720 | 0x1b8000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x38f4b720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x38f4f720 | 0x900000 |
146
+ | 43 | blk.4.attn_q.weight | 0x3984f720 | 0x6e0000 |
147
+ | 44 | blk.4.attn_v.weight | 0x39f2f720 | 0x240000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x3a16f720 | 0x1f80000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x3c0ef720 | 0x1810000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x3d8ff720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x3d903720 | 0x1810000 |
152
+ | 49 | blk.5.attn_k.weight | 0x3f113720 | 0x1b8000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x3f2cb720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x3f2cf720 | 0x900000 |
155
+ | 52 | blk.5.attn_q.weight | 0x3fbcf720 | 0x6e0000 |
156
+ | 53 | blk.5.attn_v.weight | 0x402af720 | 0x240000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x404ef720 | 0x1f80000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x4246f720 | 0x1810000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x43c7f720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x43c83720 | 0x1810000 |
161
+ | 58 | blk.6.attn_k.weight | 0x45493720 | 0x1b8000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x4564b720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x4564f720 | 0x900000 |
164
+ | 61 | blk.6.attn_q.weight | 0x45f4f720 | 0x6e0000 |
165
+ | 62 | blk.6.attn_v.weight | 0x4662f720 | 0x240000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x4686f720 | 0x1f80000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x487ef720 | 0x1810000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x49fff720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x4a003720 | 0x1810000 |
170
+ | 67 | blk.7.attn_k.weight | 0x4b813720 | 0x1b8000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x4b9cb720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x4b9cf720 | 0x900000 |
173
+ | 70 | blk.7.attn_q.weight | 0x4c2cf720 | 0x6e0000 |
174
+ | 71 | blk.7.attn_v.weight | 0x4c9af720 | 0x240000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x4cbef720 | 0x1f80000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x4eb6f720 | 0x1810000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x5037f720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x50383720 | 0x1810000 |
179
+ | 76 | blk.8.attn_k.weight | 0x51b93720 | 0x1b8000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x51d4b720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x51d4f720 | 0x900000 |
182
+ | 79 | blk.8.attn_q.weight | 0x5264f720 | 0x6e0000 |
183
+ | 80 | blk.8.attn_v.weight | 0x52d2f720 | 0x240000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x52f6f720 | 0x1f80000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x54eef720 | 0x1810000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x566ff720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x56703720 | 0x1810000 |
188
+ | 85 | blk.9.attn_k.weight | 0x57f13720 | 0x1b8000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x580cb720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x580cf720 | 0x900000 |
191
+ | 88 | blk.9.attn_q.weight | 0x589cf720 | 0x6e0000 |
192
+ | 89 | blk.9.attn_v.weight | 0x590af720 | 0x240000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x592ef720 | 0x1f80000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x5b26f720 | 0x1810000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0x5ca7f720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0x5ca83720 | 0x1810000 |
197
+ | 94 | blk.10.attn_k.weight | 0x5e293720 | 0x1b8000 |
198
+ | 95 | blk.10.attn_norm.weight | 0x5e44b720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0x5e44f720 | 0x900000 |
200
+ | 97 | blk.10.attn_q.weight | 0x5ed4f720 | 0x6e0000 |
201
+ | 98 | blk.10.attn_v.weight | 0x5f42f720 | 0x240000 |
202
+ | 99 | blk.10.ffn_down.weight | 0x5f66f720 | 0x1f80000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0x615ef720 | 0x1810000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0x62dff720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0x62e03720 | 0x1810000 |
206
+ | 103 | blk.11.attn_k.weight | 0x64613720 | 0x1b8000 |
207
+ | 104 | blk.11.attn_norm.weight | 0x647cb720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0x647cf720 | 0x900000 |
209
+ | 106 | blk.11.attn_q.weight | 0x650cf720 | 0x6e0000 |
210
+ | 107 | blk.11.attn_v.weight | 0x657af720 | 0x240000 |
211
+ | 108 | blk.11.ffn_down.weight | 0x659ef720 | 0x1f80000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0x6796f720 | 0x1810000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0x6917f720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0x69183720 | 0x1810000 |
215
+ | 112 | blk.12.attn_k.weight | 0x6a993720 | 0x1b8000 |
216
+ | 113 | blk.12.attn_norm.weight | 0x6ab4b720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0x6ab4f720 | 0x900000 |
218
+ | 115 | blk.12.attn_q.weight | 0x6b44f720 | 0x6e0000 |
219
+ | 116 | blk.12.attn_v.weight | 0x6bb2f720 | 0x240000 |
220
+ | 117 | blk.12.ffn_down.weight | 0x6bd6f720 | 0x1f80000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0x6dcef720 | 0x1810000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0x6f4ff720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0x6f503720 | 0x1810000 |
224
+ | 121 | blk.13.attn_k.weight | 0x70d13720 | 0x240000 |
225
+ | 122 | blk.13.attn_norm.weight | 0x70f53720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0x70f57720 | 0x900000 |
227
+ | 124 | blk.13.attn_q.weight | 0x71857720 | 0x900000 |
228
+ | 125 | blk.13.attn_v.weight | 0x72157720 | 0x240000 |
229
+ | 126 | blk.13.ffn_down.weight | 0x72397720 | 0x1f80000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0x74317720 | 0x1810000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0x75b27720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0x75b2b720 | 0x1810000 |
233
+ | 130 | blk.14.attn_k.weight | 0x7733b720 | 0x240000 |
234
+ | 131 | blk.14.attn_norm.weight | 0x7757b720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0x7757f720 | 0x900000 |
236
+ | 133 | blk.14.attn_q.weight | 0x77e7f720 | 0x900000 |
237
+ | 134 | blk.14.attn_v.weight | 0x7877f720 | 0x240000 |
238
+ | 135 | blk.14.ffn_down.weight | 0x789bf720 | 0x1f80000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0x7a93f720 | 0x1810000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0x7c14f720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0x7c153720 | 0x1810000 |
242
+ | 139 | blk.15.attn_k.weight | 0x7d963720 | 0x1b8000 |
243
+ | 140 | blk.15.attn_norm.weight | 0x7db1b720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0x7db1f720 | 0x900000 |
245
+ | 142 | blk.15.attn_q.weight | 0x7e41f720 | 0x6e0000 |
246
+ | 143 | blk.15.attn_v.weight | 0x7eaff720 | 0x240000 |
247
+ | 144 | blk.15.ffn_down.weight | 0x7ed3f720 | 0x1f80000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0x80cbf720 | 0x1810000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0x824cf720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0x824d3720 | 0x1810000 |
251
+ | 148 | blk.16.attn_k.weight | 0x83ce3720 | 0x240000 |
252
+ | 149 | blk.16.attn_norm.weight | 0x83f23720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0x83f27720 | 0x900000 |
254
+ | 151 | blk.16.attn_q.weight | 0x84827720 | 0x900000 |
255
+ | 152 | blk.16.attn_v.weight | 0x85127720 | 0x240000 |
256
+ | 153 | blk.16.ffn_down.weight | 0x85367720 | 0x1f80000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0x872e7720 | 0x1f80000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0x89267720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0x8926b720 | 0x1f80000 |
260
+ | 157 | blk.17.attn_k.weight | 0x8b1eb720 | 0x1b8000 |
261
+ | 158 | blk.17.attn_norm.weight | 0x8b3a3720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0x8b3a7720 | 0x900000 |
263
+ | 160 | blk.17.attn_q.weight | 0x8bca7720 | 0x6e0000 |
264
+ | 161 | blk.17.attn_v.weight | 0x8c387720 | 0x240000 |
265
+ | 162 | blk.17.ffn_down.weight | 0x8c5c7720 | 0x2680000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0x8ec47720 | 0x1f80000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0x90bc7720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0x90bcb720 | 0x1f80000 |
269
+ | 166 | blk.18.attn_k.weight | 0x92b4b720 | 0x240000 |
270
+ | 167 | blk.18.attn_norm.weight | 0x92d8b720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0x92d8f720 | 0x900000 |
272
+ | 169 | blk.18.attn_q.weight | 0x9368f720 | 0x900000 |
273
+ | 170 | blk.18.attn_v.weight | 0x93f8f720 | 0x240000 |
274
+ | 171 | blk.18.ffn_down.weight | 0x941cf720 | 0x2680000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0x9684f720 | 0x1f80000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0x987cf720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0x987d3720 | 0x1f80000 |
278
+ | 175 | blk.19.attn_k.weight | 0x9a753720 | 0x240000 |
279
+ | 176 | blk.19.attn_norm.weight | 0x9a993720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0x9a997720 | 0x900000 |
281
+ | 178 | blk.19.attn_q.weight | 0x9b297720 | 0x900000 |
282
+ | 179 | blk.19.attn_v.weight | 0x9bb97720 | 0x240000 |
283
+ | 180 | blk.19.ffn_down.weight | 0x9bdd7720 | 0x2680000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0x9e457720 | 0x1f80000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0xa03d7720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0xa03db720 | 0x1f80000 |
287
+ | 184 | blk.20.attn_k.weight | 0xa235b720 | 0x240000 |
288
+ | 185 | blk.20.attn_norm.weight | 0xa259b720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0xa259f720 | 0x900000 |
290
+ | 187 | blk.20.attn_q.weight | 0xa2e9f720 | 0x900000 |
291
+ | 188 | blk.20.attn_v.weight | 0xa379f720 | 0x240000 |
292
+ | 189 | blk.20.ffn_down.weight | 0xa39df720 | 0x2680000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0xa605f720 | 0x1f80000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0xa7fdf720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0xa7fe3720 | 0x1f80000 |
296
+ | 193 | blk.21.attn_k.weight | 0xa9f63720 | 0x240000 |
297
+ | 194 | blk.21.attn_norm.weight | 0xaa1a3720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0xaa1a7720 | 0x900000 |
299
+ | 196 | blk.21.attn_q.weight | 0xaaaa7720 | 0x900000 |
300
+ | 197 | blk.21.attn_v.weight | 0xab3a7720 | 0x240000 |
301
+ | 198 | blk.21.ffn_down.weight | 0xab5e7720 | 0x2680000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0xadc67720 | 0x1f80000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0xafbe7720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0xafbeb720 | 0x1f80000 |
305
+ | 202 | blk.22.attn_k.weight | 0xb1b6b720 | 0x240000 |
306
+ | 203 | blk.22.attn_norm.weight | 0xb1dab720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0xb1daf720 | 0x900000 |
308
+ | 205 | blk.22.attn_q.weight | 0xb26af720 | 0x900000 |
309
+ | 206 | blk.22.attn_v.weight | 0xb2faf720 | 0x240000 |
310
+ | 207 | blk.22.ffn_down.weight | 0xb31ef720 | 0x2680000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0xb586f720 | 0x1f80000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0xb77ef720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0xb77f3720 | 0x1f80000 |
314
+ | 211 | blk.23.attn_k.weight | 0xb9773720 | 0x240000 |
315
+ | 212 | blk.23.attn_norm.weight | 0xb99b3720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0xb99b7720 | 0x900000 |
317
+ | 214 | blk.23.attn_q.weight | 0xba2b7720 | 0x900000 |
318
+ | 215 | blk.23.attn_v.weight | 0xbabb7720 | 0x240000 |
319
+ | 216 | blk.23.ffn_down.weight | 0xbadf7720 | 0x2680000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0xbd477720 | 0x1f80000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0xbf3f7720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0xbf3fb720 | 0x1f80000 |
323
+ | 220 | blk.24.attn_k.weight | 0xc137b720 | 0x240000 |
324
+ | 221 | blk.24.attn_norm.weight | 0xc15bb720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0xc15bf720 | 0x900000 |
326
+ | 223 | blk.24.attn_q.weight | 0xc1ebf720 | 0x900000 |
327
+ | 224 | blk.24.attn_v.weight | 0xc27bf720 | 0x240000 |
328
+ | 225 | blk.24.ffn_down.weight | 0xc29ff720 | 0x2680000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0xc507f720 | 0x1f80000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0xc6fff720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0xc7003720 | 0x1f80000 |
332
+ | 229 | blk.25.attn_k.weight | 0xc8f83720 | 0x240000 |
333
+ | 230 | blk.25.attn_norm.weight | 0xc91c3720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0xc91c7720 | 0x900000 |
335
+ | 232 | blk.25.attn_q.weight | 0xc9ac7720 | 0x900000 |
336
+ | 233 | blk.25.attn_v.weight | 0xca3c7720 | 0x240000 |
337
+ | 234 | blk.25.ffn_down.weight | 0xca607720 | 0x2680000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0xccc87720 | 0x1f80000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0xcec07720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0xcec0b720 | 0x1f80000 |
341
+ | 238 | blk.26.attn_k.weight | 0xd0b8b720 | 0x240000 |
342
+ | 239 | blk.26.attn_norm.weight | 0xd0dcb720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0xd0dcf720 | 0x900000 |
344
+ | 241 | blk.26.attn_q.weight | 0xd16cf720 | 0x900000 |
345
+ | 242 | blk.26.attn_v.weight | 0xd1fcf720 | 0x240000 |
346
+ | 243 | blk.26.ffn_down.weight | 0xd220f720 | 0x2680000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0xd488f720 | 0x1f80000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0xd680f720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0xd6813720 | 0x1f80000 |
350
+ | 247 | blk.27.attn_k.weight | 0xd8793720 | 0x240000 |
351
+ | 248 | blk.27.attn_norm.weight | 0xd89d3720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0xd89d7720 | 0x900000 |
353
+ | 250 | blk.27.attn_q.weight | 0xd92d7720 | 0x900000 |
354
+ | 251 | blk.27.attn_v.weight | 0xd9bd7720 | 0x240000 |
355
+ | 252 | blk.27.ffn_down.weight | 0xd9e17720 | 0x2680000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0xdc497720 | 0x1f80000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0xde417720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0xde41b720 | 0x1f80000 |
359
+ | 256 | blk.28.attn_k.weight | 0xe039b720 | 0x240000 |
360
+ | 257 | blk.28.attn_norm.weight | 0xe05db720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0xe05df720 | 0x900000 |
362
+ | 259 | blk.28.attn_q.weight | 0xe0edf720 | 0x900000 |
363
+ | 260 | blk.28.attn_v.weight | 0xe17df720 | 0x240000 |
364
+ | 261 | blk.28.ffn_down.weight | 0xe1a1f720 | 0x2680000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0xe409f720 | 0x1f80000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0xe601f720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0xe6023720 | 0x1f80000 |
368
+ | 265 | blk.29.attn_k.weight | 0xe7fa3720 | 0x240000 |
369
+ | 266 | blk.29.attn_norm.weight | 0xe81e3720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0xe81e7720 | 0x900000 |
371
+ | 268 | blk.29.attn_q.weight | 0xe8ae7720 | 0x900000 |
372
+ | 269 | blk.29.attn_v.weight | 0xe93e7720 | 0x240000 |
373
+ | 270 | blk.29.ffn_down.weight | 0xe9627720 | 0x2680000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0xebca7720 | 0x1f80000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0xedc27720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0xedc2b720 | 0x1f80000 |
377
+ | 274 | blk.30.attn_k.weight | 0xefbab720 | 0x240000 |
378
+ | 275 | blk.30.attn_norm.weight | 0xefdeb720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0xefdef720 | 0x900000 |
380
+ | 277 | blk.30.attn_q.weight | 0xf06ef720 | 0x900000 |
381
+ | 278 | blk.30.attn_v.weight | 0xf0fef720 | 0x240000 |
382
+ | 279 | blk.30.ffn_down.weight | 0xf122f720 | 0x2680000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0xf38af720 | 0x1f80000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0xf582f720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0xf5833720 | 0x1f80000 |
386
+ | 283 | blk.31.attn_k.weight | 0xf77b3720 | 0x1b8000 |
387
+ | 284 | blk.31.attn_norm.weight | 0xf796b720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0xf796f720 | 0x900000 |
389
+ | 286 | blk.31.attn_q.weight | 0xf826f720 | 0x6e0000 |
390
+ | 287 | blk.31.attn_v.weight | 0xf894f720 | 0x240000 |
391
+ | 288 | blk.31.ffn_down.weight | 0xf8b8f720 | 0x2680000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0xfb20f720 | 0x1f80000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0xfd18f720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0xfd193720 | 0x1f80000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q4_K |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q3_K |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q3_K |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q4_K |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q3_K |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q3_K |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-Q5_K_M.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-Q5_K_M.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 17 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-Q5\_K\_M.gguf - GGUF Internal File Dump](#watt-tool-8b-q5_k_mgguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0x15870000 |
104
+ | 1 | output_norm.weight | 0x15fe9620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0x15fed620 | 0x100 |
106
+ | 3 | token_embd.weight | 0x15fed720 | 0xd746000 |
107
+ | 4 | blk.0.attn_k.weight | 0x23733720 | 0x240000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x23973720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x23977720 | 0xb00000 |
110
+ | 7 | blk.0.attn_q.weight | 0x24477720 | 0x900000 |
111
+ | 8 | blk.0.attn_v.weight | 0x24d77720 | 0x2c0000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x25037720 | 0x2df0000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x27e27720 | 0x1f80000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x29da7720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x29dab720 | 0x1f80000 |
116
+ | 13 | blk.1.attn_k.weight | 0x2bd2b720 | 0x240000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x2bf6b720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x2bf6f720 | 0xb00000 |
119
+ | 16 | blk.1.attn_q.weight | 0x2ca6f720 | 0x900000 |
120
+ | 17 | blk.1.attn_v.weight | 0x2d36f720 | 0x2c0000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x2d62f720 | 0x2df0000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x3041f720 | 0x1f80000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x3239f720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x323a3720 | 0x1f80000 |
125
+ | 22 | blk.2.attn_k.weight | 0x34323720 | 0x240000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x34563720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x34567720 | 0xb00000 |
128
+ | 25 | blk.2.attn_q.weight | 0x35067720 | 0x900000 |
129
+ | 26 | blk.2.attn_v.weight | 0x35967720 | 0x2c0000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x35c27720 | 0x2df0000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x38a17720 | 0x1f80000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x3a997720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x3a99b720 | 0x1f80000 |
134
+ | 31 | blk.3.attn_k.weight | 0x3c91b720 | 0x240000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x3cb5b720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x3cb5f720 | 0xb00000 |
137
+ | 34 | blk.3.attn_q.weight | 0x3d65f720 | 0x900000 |
138
+ | 35 | blk.3.attn_v.weight | 0x3df5f720 | 0x2c0000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x3e21f720 | 0x2df0000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x4100f720 | 0x1f80000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x42f8f720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x42f93720 | 0x1f80000 |
143
+ | 40 | blk.4.attn_k.weight | 0x44f13720 | 0x240000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x45153720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x45157720 | 0xb00000 |
146
+ | 43 | blk.4.attn_q.weight | 0x45c57720 | 0x900000 |
147
+ | 44 | blk.4.attn_v.weight | 0x46557720 | 0x2c0000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x46817720 | 0x2df0000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x49607720 | 0x1f80000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x4b587720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x4b58b720 | 0x1f80000 |
152
+ | 49 | blk.5.attn_k.weight | 0x4d50b720 | 0x240000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x4d74b720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x4d74f720 | 0xb00000 |
155
+ | 52 | blk.5.attn_q.weight | 0x4e24f720 | 0x900000 |
156
+ | 53 | blk.5.attn_v.weight | 0x4eb4f720 | 0x2c0000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x4ee0f720 | 0x2df0000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x51bff720 | 0x1f80000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x53b7f720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x53b83720 | 0x1f80000 |
161
+ | 58 | blk.6.attn_k.weight | 0x55b03720 | 0x240000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x55d43720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x55d47720 | 0xb00000 |
164
+ | 61 | blk.6.attn_q.weight | 0x56847720 | 0x900000 |
165
+ | 62 | blk.6.attn_v.weight | 0x57147720 | 0x2c0000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x57407720 | 0x2df0000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x5a1f7720 | 0x1f80000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x5c177720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x5c17b720 | 0x1f80000 |
170
+ | 67 | blk.7.attn_k.weight | 0x5e0fb720 | 0x240000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x5e33b720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x5e33f720 | 0xb00000 |
173
+ | 70 | blk.7.attn_q.weight | 0x5ee3f720 | 0x900000 |
174
+ | 71 | blk.7.attn_v.weight | 0x5f73f720 | 0x2c0000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x5f9ff720 | 0x2df0000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x627ef720 | 0x1f80000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x6476f720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x64773720 | 0x1f80000 |
179
+ | 76 | blk.8.attn_k.weight | 0x666f3720 | 0x240000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x66933720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x66937720 | 0xb00000 |
182
+ | 79 | blk.8.attn_q.weight | 0x67437720 | 0x900000 |
183
+ | 80 | blk.8.attn_v.weight | 0x67d37720 | 0x2c0000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x67ff7720 | 0x2df0000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x6ade7720 | 0x1f80000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x6cd67720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x6cd6b720 | 0x1f80000 |
188
+ | 85 | blk.9.attn_k.weight | 0x6eceb720 | 0x240000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x6ef2b720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x6ef2f720 | 0xb00000 |
191
+ | 88 | blk.9.attn_q.weight | 0x6fa2f720 | 0x900000 |
192
+ | 89 | blk.9.attn_v.weight | 0x7032f720 | 0x2c0000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x705ef720 | 0x2df0000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x733df720 | 0x1f80000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0x7535f720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0x75363720 | 0x1f80000 |
197
+ | 94 | blk.10.attn_k.weight | 0x772e3720 | 0x240000 |
198
+ | 95 | blk.10.attn_norm.weight | 0x77523720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0x77527720 | 0xb00000 |
200
+ | 97 | blk.10.attn_q.weight | 0x78027720 | 0x900000 |
201
+ | 98 | blk.10.attn_v.weight | 0x78927720 | 0x2c0000 |
202
+ | 99 | blk.10.ffn_down.weight | 0x78be7720 | 0x2df0000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0x7b9d7720 | 0x1f80000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0x7d957720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0x7d95b720 | 0x1f80000 |
206
+ | 103 | blk.11.attn_k.weight | 0x7f8db720 | 0x240000 |
207
+ | 104 | blk.11.attn_norm.weight | 0x7fb1b720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0x7fb1f720 | 0xb00000 |
209
+ | 106 | blk.11.attn_q.weight | 0x8061f720 | 0x900000 |
210
+ | 107 | blk.11.attn_v.weight | 0x80f1f720 | 0x2c0000 |
211
+ | 108 | blk.11.ffn_down.weight | 0x811df720 | 0x2df0000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0x83fcf720 | 0x1f80000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0x85f4f720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0x85f53720 | 0x1f80000 |
215
+ | 112 | blk.12.attn_k.weight | 0x87ed3720 | 0x240000 |
216
+ | 113 | blk.12.attn_norm.weight | 0x88113720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0x88117720 | 0xb00000 |
218
+ | 115 | blk.12.attn_q.weight | 0x88c17720 | 0x900000 |
219
+ | 116 | blk.12.attn_v.weight | 0x89517720 | 0x2c0000 |
220
+ | 117 | blk.12.ffn_down.weight | 0x897d7720 | 0x2df0000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0x8c5c7720 | 0x1f80000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0x8e547720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0x8e54b720 | 0x1f80000 |
224
+ | 121 | blk.13.attn_k.weight | 0x904cb720 | 0x2c0000 |
225
+ | 122 | blk.13.attn_norm.weight | 0x9078b720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0x9078f720 | 0xb00000 |
227
+ | 124 | blk.13.attn_q.weight | 0x9128f720 | 0xb00000 |
228
+ | 125 | blk.13.attn_v.weight | 0x91d8f720 | 0x348000 |
229
+ | 126 | blk.13.ffn_down.weight | 0x920d7720 | 0x2df0000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0x94ec7720 | 0x1f80000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0x96e47720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0x96e4b720 | 0x1f80000 |
233
+ | 130 | blk.14.attn_k.weight | 0x98dcb720 | 0x2c0000 |
234
+ | 131 | blk.14.attn_norm.weight | 0x9908b720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0x9908f720 | 0xb00000 |
236
+ | 133 | blk.14.attn_q.weight | 0x99b8f720 | 0xb00000 |
237
+ | 134 | blk.14.attn_v.weight | 0x9a68f720 | 0x348000 |
238
+ | 135 | blk.14.ffn_down.weight | 0x9a9d7720 | 0x2df0000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0x9d7c7720 | 0x1f80000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0x9f747720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0x9f74b720 | 0x1f80000 |
242
+ | 139 | blk.15.attn_k.weight | 0xa16cb720 | 0x240000 |
243
+ | 140 | blk.15.attn_norm.weight | 0xa190b720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0xa190f720 | 0xb00000 |
245
+ | 142 | blk.15.attn_q.weight | 0xa240f720 | 0x900000 |
246
+ | 143 | blk.15.attn_v.weight | 0xa2d0f720 | 0x2c0000 |
247
+ | 144 | blk.15.ffn_down.weight | 0xa2fcf720 | 0x2df0000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0xa5dbf720 | 0x1f80000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0xa7d3f720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0xa7d43720 | 0x1f80000 |
251
+ | 148 | blk.16.attn_k.weight | 0xa9cc3720 | 0x2c0000 |
252
+ | 149 | blk.16.attn_norm.weight | 0xa9f83720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0xa9f87720 | 0xb00000 |
254
+ | 151 | blk.16.attn_q.weight | 0xaaa87720 | 0xb00000 |
255
+ | 152 | blk.16.attn_v.weight | 0xab587720 | 0x348000 |
256
+ | 153 | blk.16.ffn_down.weight | 0xab8cf720 | 0x2df0000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0xae6bf720 | 0x2680000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0xb0d3f720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0xb0d43720 | 0x2680000 |
260
+ | 157 | blk.17.attn_k.weight | 0xb33c3720 | 0x240000 |
261
+ | 158 | blk.17.attn_norm.weight | 0xb3603720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0xb3607720 | 0xb00000 |
263
+ | 160 | blk.17.attn_q.weight | 0xb4107720 | 0x900000 |
264
+ | 161 | blk.17.attn_v.weight | 0xb4a07720 | 0x2c0000 |
265
+ | 162 | blk.17.ffn_down.weight | 0xb4cc7720 | 0x2df0000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0xb7ab7720 | 0x2680000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0xba137720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0xba13b720 | 0x2680000 |
269
+ | 166 | blk.18.attn_k.weight | 0xbc7bb720 | 0x2c0000 |
270
+ | 167 | blk.18.attn_norm.weight | 0xbca7b720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0xbca7f720 | 0xb00000 |
272
+ | 169 | blk.18.attn_q.weight | 0xbd57f720 | 0xb00000 |
273
+ | 170 | blk.18.attn_v.weight | 0xbe07f720 | 0x348000 |
274
+ | 171 | blk.18.ffn_down.weight | 0xbe3c7720 | 0x2df0000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0xc11b7720 | 0x2680000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0xc3837720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0xc383b720 | 0x2680000 |
278
+ | 175 | blk.19.attn_k.weight | 0xc5ebb720 | 0x2c0000 |
279
+ | 176 | blk.19.attn_norm.weight | 0xc617b720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0xc617f720 | 0xb00000 |
281
+ | 178 | blk.19.attn_q.weight | 0xc6c7f720 | 0xb00000 |
282
+ | 179 | blk.19.attn_v.weight | 0xc777f720 | 0x348000 |
283
+ | 180 | blk.19.ffn_down.weight | 0xc7ac7720 | 0x2df0000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0xca8b7720 | 0x2680000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0xccf37720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0xccf3b720 | 0x2680000 |
287
+ | 184 | blk.20.attn_k.weight | 0xcf5bb720 | 0x2c0000 |
288
+ | 185 | blk.20.attn_norm.weight | 0xcf87b720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0xcf87f720 | 0xb00000 |
290
+ | 187 | blk.20.attn_q.weight | 0xd037f720 | 0xb00000 |
291
+ | 188 | blk.20.attn_v.weight | 0xd0e7f720 | 0x348000 |
292
+ | 189 | blk.20.ffn_down.weight | 0xd11c7720 | 0x2df0000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0xd3fb7720 | 0x2680000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0xd6637720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0xd663b720 | 0x2680000 |
296
+ | 193 | blk.21.attn_k.weight | 0xd8cbb720 | 0x2c0000 |
297
+ | 194 | blk.21.attn_norm.weight | 0xd8f7b720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0xd8f7f720 | 0xb00000 |
299
+ | 196 | blk.21.attn_q.weight | 0xd9a7f720 | 0xb00000 |
300
+ | 197 | blk.21.attn_v.weight | 0xda57f720 | 0x348000 |
301
+ | 198 | blk.21.ffn_down.weight | 0xda8c7720 | 0x2df0000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0xdd6b7720 | 0x2680000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0xdfd37720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0xdfd3b720 | 0x2680000 |
305
+ | 202 | blk.22.attn_k.weight | 0xe23bb720 | 0x2c0000 |
306
+ | 203 | blk.22.attn_norm.weight | 0xe267b720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0xe267f720 | 0xb00000 |
308
+ | 205 | blk.22.attn_q.weight | 0xe317f720 | 0xb00000 |
309
+ | 206 | blk.22.attn_v.weight | 0xe3c7f720 | 0x348000 |
310
+ | 207 | blk.22.ffn_down.weight | 0xe3fc7720 | 0x2df0000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0xe6db7720 | 0x2680000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0xe9437720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0xe943b720 | 0x2680000 |
314
+ | 211 | blk.23.attn_k.weight | 0xebabb720 | 0x2c0000 |
315
+ | 212 | blk.23.attn_norm.weight | 0xebd7b720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0xebd7f720 | 0xb00000 |
317
+ | 214 | blk.23.attn_q.weight | 0xec87f720 | 0xb00000 |
318
+ | 215 | blk.23.attn_v.weight | 0xed37f720 | 0x348000 |
319
+ | 216 | blk.23.ffn_down.weight | 0xed6c7720 | 0x2df0000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0xf04b7720 | 0x2680000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0xf2b37720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0xf2b3b720 | 0x2680000 |
323
+ | 220 | blk.24.attn_k.weight | 0xf51bb720 | 0x2c0000 |
324
+ | 221 | blk.24.attn_norm.weight | 0xf547b720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0xf547f720 | 0xb00000 |
326
+ | 223 | blk.24.attn_q.weight | 0xf5f7f720 | 0xb00000 |
327
+ | 224 | blk.24.attn_v.weight | 0xf6a7f720 | 0x348000 |
328
+ | 225 | blk.24.ffn_down.weight | 0xf6dc7720 | 0x2df0000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0xf9bb7720 | 0x2680000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0xfc237720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0xfc23b720 | 0x2680000 |
332
+ | 229 | blk.25.attn_k.weight | 0xfe8bb720 | 0x2c0000 |
333
+ | 230 | blk.25.attn_norm.weight | 0xfeb7b720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0xfeb7f720 | 0xb00000 |
335
+ | 232 | blk.25.attn_q.weight | 0xff67f720 | 0xb00000 |
336
+ | 233 | blk.25.attn_v.weight | 0x10017f720 | 0x348000 |
337
+ | 234 | blk.25.ffn_down.weight | 0x1004c7720 | 0x2df0000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0x1032b7720 | 0x2680000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0x105937720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0x10593b720 | 0x2680000 |
341
+ | 238 | blk.26.attn_k.weight | 0x107fbb720 | 0x2c0000 |
342
+ | 239 | blk.26.attn_norm.weight | 0x10827b720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0x10827f720 | 0xb00000 |
344
+ | 241 | blk.26.attn_q.weight | 0x108d7f720 | 0xb00000 |
345
+ | 242 | blk.26.attn_v.weight | 0x10987f720 | 0x348000 |
346
+ | 243 | blk.26.ffn_down.weight | 0x109bc7720 | 0x2df0000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0x10c9b7720 | 0x2680000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0x10f037720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0x10f03b720 | 0x2680000 |
350
+ | 247 | blk.27.attn_k.weight | 0x1116bb720 | 0x2c0000 |
351
+ | 248 | blk.27.attn_norm.weight | 0x11197b720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0x11197f720 | 0xb00000 |
353
+ | 250 | blk.27.attn_q.weight | 0x11247f720 | 0xb00000 |
354
+ | 251 | blk.27.attn_v.weight | 0x112f7f720 | 0x348000 |
355
+ | 252 | blk.27.ffn_down.weight | 0x1132c7720 | 0x2df0000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0x1160b7720 | 0x2680000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0x118737720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0x11873b720 | 0x2680000 |
359
+ | 256 | blk.28.attn_k.weight | 0x11adbb720 | 0x2c0000 |
360
+ | 257 | blk.28.attn_norm.weight | 0x11b07b720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0x11b07f720 | 0xb00000 |
362
+ | 259 | blk.28.attn_q.weight | 0x11bb7f720 | 0xb00000 |
363
+ | 260 | blk.28.attn_v.weight | 0x11c67f720 | 0x348000 |
364
+ | 261 | blk.28.ffn_down.weight | 0x11c9c7720 | 0x2df0000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0x11f7b7720 | 0x2680000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0x121e37720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0x121e3b720 | 0x2680000 |
368
+ | 265 | blk.29.attn_k.weight | 0x1244bb720 | 0x2c0000 |
369
+ | 266 | blk.29.attn_norm.weight | 0x12477b720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0x12477f720 | 0xb00000 |
371
+ | 268 | blk.29.attn_q.weight | 0x12527f720 | 0xb00000 |
372
+ | 269 | blk.29.attn_v.weight | 0x125d7f720 | 0x348000 |
373
+ | 270 | blk.29.ffn_down.weight | 0x1260c7720 | 0x2df0000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0x128eb7720 | 0x2680000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0x12b537720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0x12b53b720 | 0x2680000 |
377
+ | 274 | blk.30.attn_k.weight | 0x12dbbb720 | 0x2c0000 |
378
+ | 275 | blk.30.attn_norm.weight | 0x12de7b720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0x12de7f720 | 0xb00000 |
380
+ | 277 | blk.30.attn_q.weight | 0x12e97f720 | 0xb00000 |
381
+ | 278 | blk.30.attn_v.weight | 0x12f47f720 | 0x348000 |
382
+ | 279 | blk.30.ffn_down.weight | 0x12f7c7720 | 0x2df0000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0x1325b7720 | 0x2680000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0x134c37720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0x134c3b720 | 0x2680000 |
386
+ | 283 | blk.31.attn_k.weight | 0x1372bb720 | 0x240000 |
387
+ | 284 | blk.31.attn_norm.weight | 0x1374fb720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0x1374ff720 | 0xb00000 |
389
+ | 286 | blk.31.attn_q.weight | 0x137fff720 | 0x900000 |
390
+ | 287 | blk.31.attn_v.weight | 0x1388ff720 | 0x2c0000 |
391
+ | 288 | blk.31.ffn_down.weight | 0x138bbf720 | 0x2df0000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0x13b9af720 | 0x2680000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0x13e02f720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0x13e033720 | 0x2680000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q5_K |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q3_K |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-Q5_K_S.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-Q5_K_S.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 16 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-Q5\_K\_S.gguf - GGUF Internal File Dump](#watt-tool-8b-q5_k_sgguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0x15870000 |
104
+ | 1 | output_norm.weight | 0x15fe9620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0x15fed620 | 0x100 |
106
+ | 3 | token_embd.weight | 0x15fed720 | 0xd746000 |
107
+ | 4 | blk.0.attn_k.weight | 0x23733720 | 0x240000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x23973720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x23977720 | 0xb00000 |
110
+ | 7 | blk.0.attn_q.weight | 0x24477720 | 0x900000 |
111
+ | 8 | blk.0.attn_v.weight | 0x24d77720 | 0x2c0000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x25037720 | 0x2680000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x276b7720 | 0x1f80000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x29637720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x2963b720 | 0x1f80000 |
116
+ | 13 | blk.1.attn_k.weight | 0x2b5bb720 | 0x240000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x2b7fb720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x2b7ff720 | 0xb00000 |
119
+ | 16 | blk.1.attn_q.weight | 0x2c2ff720 | 0x900000 |
120
+ | 17 | blk.1.attn_v.weight | 0x2cbff720 | 0x2c0000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x2cebf720 | 0x2df0000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x2fcaf720 | 0x1f80000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x31c2f720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x31c33720 | 0x1f80000 |
125
+ | 22 | blk.2.attn_k.weight | 0x33bb3720 | 0x240000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x33df3720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x33df7720 | 0xb00000 |
128
+ | 25 | blk.2.attn_q.weight | 0x348f7720 | 0x900000 |
129
+ | 26 | blk.2.attn_v.weight | 0x351f7720 | 0x2c0000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x354b7720 | 0x2680000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x37b37720 | 0x1f80000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x39ab7720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x39abb720 | 0x1f80000 |
134
+ | 31 | blk.3.attn_k.weight | 0x3ba3b720 | 0x240000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x3bc7b720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x3bc7f720 | 0xb00000 |
137
+ | 34 | blk.3.attn_q.weight | 0x3c77f720 | 0x900000 |
138
+ | 35 | blk.3.attn_v.weight | 0x3d07f720 | 0x2c0000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x3d33f720 | 0x2680000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x3f9bf720 | 0x1f80000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x4193f720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x41943720 | 0x1f80000 |
143
+ | 40 | blk.4.attn_k.weight | 0x438c3720 | 0x240000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x43b03720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x43b07720 | 0xb00000 |
146
+ | 43 | blk.4.attn_q.weight | 0x44607720 | 0x900000 |
147
+ | 44 | blk.4.attn_v.weight | 0x44f07720 | 0x2c0000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x451c7720 | 0x2680000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x47847720 | 0x1f80000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x497c7720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x497cb720 | 0x1f80000 |
152
+ | 49 | blk.5.attn_k.weight | 0x4b74b720 | 0x240000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x4b98b720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x4b98f720 | 0xb00000 |
155
+ | 52 | blk.5.attn_q.weight | 0x4c48f720 | 0x900000 |
156
+ | 53 | blk.5.attn_v.weight | 0x4cd8f720 | 0x2c0000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x4d04f720 | 0x2680000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x4f6cf720 | 0x1f80000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x5164f720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x51653720 | 0x1f80000 |
161
+ | 58 | blk.6.attn_k.weight | 0x535d3720 | 0x240000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x53813720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x53817720 | 0xb00000 |
164
+ | 61 | blk.6.attn_q.weight | 0x54317720 | 0x900000 |
165
+ | 62 | blk.6.attn_v.weight | 0x54c17720 | 0x2c0000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x54ed7720 | 0x2680000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x57557720 | 0x1f80000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x594d7720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x594db720 | 0x1f80000 |
170
+ | 67 | blk.7.attn_k.weight | 0x5b45b720 | 0x240000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x5b69b720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x5b69f720 | 0xb00000 |
173
+ | 70 | blk.7.attn_q.weight | 0x5c19f720 | 0x900000 |
174
+ | 71 | blk.7.attn_v.weight | 0x5ca9f720 | 0x2c0000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x5cd5f720 | 0x2680000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x5f3df720 | 0x1f80000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x6135f720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x61363720 | 0x1f80000 |
179
+ | 76 | blk.8.attn_k.weight | 0x632e3720 | 0x240000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x63523720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x63527720 | 0xb00000 |
182
+ | 79 | blk.8.attn_q.weight | 0x64027720 | 0x900000 |
183
+ | 80 | blk.8.attn_v.weight | 0x64927720 | 0x2c0000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x64be7720 | 0x2680000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x67267720 | 0x1f80000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x691e7720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x691eb720 | 0x1f80000 |
188
+ | 85 | blk.9.attn_k.weight | 0x6b16b720 | 0x240000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x6b3ab720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x6b3af720 | 0xb00000 |
191
+ | 88 | blk.9.attn_q.weight | 0x6beaf720 | 0x900000 |
192
+ | 89 | blk.9.attn_v.weight | 0x6c7af720 | 0x2c0000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x6ca6f720 | 0x2680000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x6f0ef720 | 0x1f80000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0x7106f720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0x71073720 | 0x1f80000 |
197
+ | 94 | blk.10.attn_k.weight | 0x72ff3720 | 0x240000 |
198
+ | 95 | blk.10.attn_norm.weight | 0x73233720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0x73237720 | 0xb00000 |
200
+ | 97 | blk.10.attn_q.weight | 0x73d37720 | 0x900000 |
201
+ | 98 | blk.10.attn_v.weight | 0x74637720 | 0x2c0000 |
202
+ | 99 | blk.10.ffn_down.weight | 0x748f7720 | 0x2680000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0x76f77720 | 0x1f80000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0x78ef7720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0x78efb720 | 0x1f80000 |
206
+ | 103 | blk.11.attn_k.weight | 0x7ae7b720 | 0x240000 |
207
+ | 104 | blk.11.attn_norm.weight | 0x7b0bb720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0x7b0bf720 | 0xb00000 |
209
+ | 106 | blk.11.attn_q.weight | 0x7bbbf720 | 0x900000 |
210
+ | 107 | blk.11.attn_v.weight | 0x7c4bf720 | 0x2c0000 |
211
+ | 108 | blk.11.ffn_down.weight | 0x7c77f720 | 0x2680000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0x7edff720 | 0x1f80000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0x80d7f720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0x80d83720 | 0x1f80000 |
215
+ | 112 | blk.12.attn_k.weight | 0x82d03720 | 0x240000 |
216
+ | 113 | blk.12.attn_norm.weight | 0x82f43720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0x82f47720 | 0xb00000 |
218
+ | 115 | blk.12.attn_q.weight | 0x83a47720 | 0x900000 |
219
+ | 116 | blk.12.attn_v.weight | 0x84347720 | 0x2c0000 |
220
+ | 117 | blk.12.ffn_down.weight | 0x84607720 | 0x2680000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0x86c87720 | 0x1f80000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0x88c07720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0x88c0b720 | 0x1f80000 |
224
+ | 121 | blk.13.attn_k.weight | 0x8ab8b720 | 0x2c0000 |
225
+ | 122 | blk.13.attn_norm.weight | 0x8ae4b720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0x8ae4f720 | 0xb00000 |
227
+ | 124 | blk.13.attn_q.weight | 0x8b94f720 | 0xb00000 |
228
+ | 125 | blk.13.attn_v.weight | 0x8c44f720 | 0x2c0000 |
229
+ | 126 | blk.13.ffn_down.weight | 0x8c70f720 | 0x2680000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0x8ed8f720 | 0x1f80000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0x90d0f720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0x90d13720 | 0x1f80000 |
233
+ | 130 | blk.14.attn_k.weight | 0x92c93720 | 0x2c0000 |
234
+ | 131 | blk.14.attn_norm.weight | 0x92f53720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0x92f57720 | 0xb00000 |
236
+ | 133 | blk.14.attn_q.weight | 0x93a57720 | 0xb00000 |
237
+ | 134 | blk.14.attn_v.weight | 0x94557720 | 0x2c0000 |
238
+ | 135 | blk.14.ffn_down.weight | 0x94817720 | 0x2680000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0x96e97720 | 0x1f80000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0x98e17720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0x98e1b720 | 0x1f80000 |
242
+ | 139 | blk.15.attn_k.weight | 0x9ad9b720 | 0x240000 |
243
+ | 140 | blk.15.attn_norm.weight | 0x9afdb720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0x9afdf720 | 0xb00000 |
245
+ | 142 | blk.15.attn_q.weight | 0x9badf720 | 0x900000 |
246
+ | 143 | blk.15.attn_v.weight | 0x9c3df720 | 0x2c0000 |
247
+ | 144 | blk.15.ffn_down.weight | 0x9c69f720 | 0x2680000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0x9ed1f720 | 0x1f80000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0xa0c9f720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0xa0ca3720 | 0x1f80000 |
251
+ | 148 | blk.16.attn_k.weight | 0xa2c23720 | 0x2c0000 |
252
+ | 149 | blk.16.attn_norm.weight | 0xa2ee3720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0xa2ee7720 | 0xb00000 |
254
+ | 151 | blk.16.attn_q.weight | 0xa39e7720 | 0xb00000 |
255
+ | 152 | blk.16.attn_v.weight | 0xa44e7720 | 0x2c0000 |
256
+ | 153 | blk.16.ffn_down.weight | 0xa47a7720 | 0x2680000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0xa6e27720 | 0x2680000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0xa94a7720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0xa94ab720 | 0x2680000 |
260
+ | 157 | blk.17.attn_k.weight | 0xabb2b720 | 0x240000 |
261
+ | 158 | blk.17.attn_norm.weight | 0xabd6b720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0xabd6f720 | 0xb00000 |
263
+ | 160 | blk.17.attn_q.weight | 0xac86f720 | 0x900000 |
264
+ | 161 | blk.17.attn_v.weight | 0xad16f720 | 0x2c0000 |
265
+ | 162 | blk.17.ffn_down.weight | 0xad42f720 | 0x2df0000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0xb021f720 | 0x2680000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0xb289f720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0xb28a3720 | 0x2680000 |
269
+ | 166 | blk.18.attn_k.weight | 0xb4f23720 | 0x2c0000 |
270
+ | 167 | blk.18.attn_norm.weight | 0xb51e3720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0xb51e7720 | 0xb00000 |
272
+ | 169 | blk.18.attn_q.weight | 0xb5ce7720 | 0xb00000 |
273
+ | 170 | blk.18.attn_v.weight | 0xb67e7720 | 0x2c0000 |
274
+ | 171 | blk.18.ffn_down.weight | 0xb6aa7720 | 0x2df0000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0xb9897720 | 0x2680000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0xbbf17720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0xbbf1b720 | 0x2680000 |
278
+ | 175 | blk.19.attn_k.weight | 0xbe59b720 | 0x2c0000 |
279
+ | 176 | blk.19.attn_norm.weight | 0xbe85b720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0xbe85f720 | 0xb00000 |
281
+ | 178 | blk.19.attn_q.weight | 0xbf35f720 | 0xb00000 |
282
+ | 179 | blk.19.attn_v.weight | 0xbfe5f720 | 0x2c0000 |
283
+ | 180 | blk.19.ffn_down.weight | 0xc011f720 | 0x2df0000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0xc2f0f720 | 0x2680000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0xc558f720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0xc5593720 | 0x2680000 |
287
+ | 184 | blk.20.attn_k.weight | 0xc7c13720 | 0x2c0000 |
288
+ | 185 | blk.20.attn_norm.weight | 0xc7ed3720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0xc7ed7720 | 0xb00000 |
290
+ | 187 | blk.20.attn_q.weight | 0xc89d7720 | 0xb00000 |
291
+ | 188 | blk.20.attn_v.weight | 0xc94d7720 | 0x2c0000 |
292
+ | 189 | blk.20.ffn_down.weight | 0xc9797720 | 0x2df0000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0xcc587720 | 0x2680000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0xcec07720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0xcec0b720 | 0x2680000 |
296
+ | 193 | blk.21.attn_k.weight | 0xd128b720 | 0x2c0000 |
297
+ | 194 | blk.21.attn_norm.weight | 0xd154b720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0xd154f720 | 0xb00000 |
299
+ | 196 | blk.21.attn_q.weight | 0xd204f720 | 0xb00000 |
300
+ | 197 | blk.21.attn_v.weight | 0xd2b4f720 | 0x2c0000 |
301
+ | 198 | blk.21.ffn_down.weight | 0xd2e0f720 | 0x2df0000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0xd5bff720 | 0x2680000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0xd827f720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0xd8283720 | 0x2680000 |
305
+ | 202 | blk.22.attn_k.weight | 0xda903720 | 0x2c0000 |
306
+ | 203 | blk.22.attn_norm.weight | 0xdabc3720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0xdabc7720 | 0xb00000 |
308
+ | 205 | blk.22.attn_q.weight | 0xdb6c7720 | 0xb00000 |
309
+ | 206 | blk.22.attn_v.weight | 0xdc1c7720 | 0x2c0000 |
310
+ | 207 | blk.22.ffn_down.weight | 0xdc487720 | 0x2df0000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0xdf277720 | 0x2680000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0xe18f7720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0xe18fb720 | 0x2680000 |
314
+ | 211 | blk.23.attn_k.weight | 0xe3f7b720 | 0x2c0000 |
315
+ | 212 | blk.23.attn_norm.weight | 0xe423b720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0xe423f720 | 0xb00000 |
317
+ | 214 | blk.23.attn_q.weight | 0xe4d3f720 | 0xb00000 |
318
+ | 215 | blk.23.attn_v.weight | 0xe583f720 | 0x2c0000 |
319
+ | 216 | blk.23.ffn_down.weight | 0xe5aff720 | 0x2df0000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0xe88ef720 | 0x2680000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0xeaf6f720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0xeaf73720 | 0x2680000 |
323
+ | 220 | blk.24.attn_k.weight | 0xed5f3720 | 0x2c0000 |
324
+ | 221 | blk.24.attn_norm.weight | 0xed8b3720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0xed8b7720 | 0xb00000 |
326
+ | 223 | blk.24.attn_q.weight | 0xee3b7720 | 0xb00000 |
327
+ | 224 | blk.24.attn_v.weight | 0xeeeb7720 | 0x2c0000 |
328
+ | 225 | blk.24.ffn_down.weight | 0xef177720 | 0x2df0000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0xf1f67720 | 0x2680000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0xf45e7720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0xf45eb720 | 0x2680000 |
332
+ | 229 | blk.25.attn_k.weight | 0xf6c6b720 | 0x2c0000 |
333
+ | 230 | blk.25.attn_norm.weight | 0xf6f2b720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0xf6f2f720 | 0xb00000 |
335
+ | 232 | blk.25.attn_q.weight | 0xf7a2f720 | 0xb00000 |
336
+ | 233 | blk.25.attn_v.weight | 0xf852f720 | 0x2c0000 |
337
+ | 234 | blk.25.ffn_down.weight | 0xf87ef720 | 0x2df0000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0xfb5df720 | 0x2680000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0xfdc5f720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0xfdc63720 | 0x2680000 |
341
+ | 238 | blk.26.attn_k.weight | 0x1002e3720 | 0x2c0000 |
342
+ | 239 | blk.26.attn_norm.weight | 0x1005a3720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0x1005a7720 | 0xb00000 |
344
+ | 241 | blk.26.attn_q.weight | 0x1010a7720 | 0xb00000 |
345
+ | 242 | blk.26.attn_v.weight | 0x101ba7720 | 0x2c0000 |
346
+ | 243 | blk.26.ffn_down.weight | 0x101e67720 | 0x2df0000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0x104c57720 | 0x2680000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0x1072d7720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0x1072db720 | 0x2680000 |
350
+ | 247 | blk.27.attn_k.weight | 0x10995b720 | 0x2c0000 |
351
+ | 248 | blk.27.attn_norm.weight | 0x109c1b720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0x109c1f720 | 0xb00000 |
353
+ | 250 | blk.27.attn_q.weight | 0x10a71f720 | 0xb00000 |
354
+ | 251 | blk.27.attn_v.weight | 0x10b21f720 | 0x2c0000 |
355
+ | 252 | blk.27.ffn_down.weight | 0x10b4df720 | 0x2df0000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0x10e2cf720 | 0x2680000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0x11094f720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0x110953720 | 0x2680000 |
359
+ | 256 | blk.28.attn_k.weight | 0x112fd3720 | 0x2c0000 |
360
+ | 257 | blk.28.attn_norm.weight | 0x113293720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0x113297720 | 0xb00000 |
362
+ | 259 | blk.28.attn_q.weight | 0x113d97720 | 0xb00000 |
363
+ | 260 | blk.28.attn_v.weight | 0x114897720 | 0x2c0000 |
364
+ | 261 | blk.28.ffn_down.weight | 0x114b57720 | 0x2df0000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0x117947720 | 0x2680000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0x119fc7720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0x119fcb720 | 0x2680000 |
368
+ | 265 | blk.29.attn_k.weight | 0x11c64b720 | 0x2c0000 |
369
+ | 266 | blk.29.attn_norm.weight | 0x11c90b720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0x11c90f720 | 0xb00000 |
371
+ | 268 | blk.29.attn_q.weight | 0x11d40f720 | 0xb00000 |
372
+ | 269 | blk.29.attn_v.weight | 0x11df0f720 | 0x2c0000 |
373
+ | 270 | blk.29.ffn_down.weight | 0x11e1cf720 | 0x2df0000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0x120fbf720 | 0x2680000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0x12363f720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0x123643720 | 0x2680000 |
377
+ | 274 | blk.30.attn_k.weight | 0x125cc3720 | 0x2c0000 |
378
+ | 275 | blk.30.attn_norm.weight | 0x125f83720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0x125f87720 | 0xb00000 |
380
+ | 277 | blk.30.attn_q.weight | 0x126a87720 | 0xb00000 |
381
+ | 278 | blk.30.attn_v.weight | 0x127587720 | 0x2c0000 |
382
+ | 279 | blk.30.ffn_down.weight | 0x127847720 | 0x2df0000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0x12a637720 | 0x2680000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0x12ccb7720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0x12ccbb720 | 0x2680000 |
386
+ | 283 | blk.31.attn_k.weight | 0x12f33b720 | 0x240000 |
387
+ | 284 | blk.31.attn_norm.weight | 0x12f57b720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0x12f57f720 | 0xb00000 |
389
+ | 286 | blk.31.attn_q.weight | 0x13007f720 | 0x900000 |
390
+ | 287 | blk.31.attn_v.weight | 0x13097f720 | 0x2c0000 |
391
+ | 288 | blk.31.ffn_down.weight | 0x130c3f720 | 0x2df0000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0x133a2f720 | 0x2680000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0x1360af720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0x1360b3720 | 0x2680000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q5_K |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q3_K |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q4_K |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q5_K |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q4_K |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q4_K |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q6_K |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-Q6_K.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-Q6_K.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 18 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-Q6\_K.gguf - GGUF Internal File Dump](#watt-tool-8b-q6_kgguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0x19afa000 |
104
+ | 1 | output_norm.weight | 0x1a273620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0x1a277620 | 0x100 |
106
+ | 3 | token_embd.weight | 0x1a277720 | 0xd746000 |
107
+ | 4 | blk.0.attn_k.weight | 0x279bd720 | 0x2c0000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x27c7d720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x27c81720 | 0xd20000 |
110
+ | 7 | blk.0.attn_q.weight | 0x289a1720 | 0xb00000 |
111
+ | 8 | blk.0.attn_v.weight | 0x294a1720 | 0x348000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x297e9720 | 0x3b80000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x2d369720 | 0x2680000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x2f9e9720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x2f9ed720 | 0x2680000 |
116
+ | 13 | blk.1.attn_k.weight | 0x3206d720 | 0x2c0000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x3232d720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x32331720 | 0xd20000 |
119
+ | 16 | blk.1.attn_q.weight | 0x33051720 | 0xb00000 |
120
+ | 17 | blk.1.attn_v.weight | 0x33b51720 | 0x348000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x33e99720 | 0x3b80000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x37a19720 | 0x2680000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x3a099720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x3a09d720 | 0x2680000 |
125
+ | 22 | blk.2.attn_k.weight | 0x3c71d720 | 0x2c0000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x3c9dd720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x3c9e1720 | 0xd20000 |
128
+ | 25 | blk.2.attn_q.weight | 0x3d701720 | 0xb00000 |
129
+ | 26 | blk.2.attn_v.weight | 0x3e201720 | 0x348000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x3e549720 | 0x3b80000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x420c9720 | 0x2680000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x44749720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x4474d720 | 0x2680000 |
134
+ | 31 | blk.3.attn_k.weight | 0x46dcd720 | 0x2c0000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x4708d720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x47091720 | 0xd20000 |
137
+ | 34 | blk.3.attn_q.weight | 0x47db1720 | 0xb00000 |
138
+ | 35 | blk.3.attn_v.weight | 0x488b1720 | 0x348000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x48bf9720 | 0x3b80000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x4c779720 | 0x2680000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x4edf9720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x4edfd720 | 0x2680000 |
143
+ | 40 | blk.4.attn_k.weight | 0x5147d720 | 0x2c0000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x5173d720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x51741720 | 0xd20000 |
146
+ | 43 | blk.4.attn_q.weight | 0x52461720 | 0xb00000 |
147
+ | 44 | blk.4.attn_v.weight | 0x52f61720 | 0x348000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x532a9720 | 0x3b80000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x56e29720 | 0x2680000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x594a9720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x594ad720 | 0x2680000 |
152
+ | 49 | blk.5.attn_k.weight | 0x5bb2d720 | 0x2c0000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x5bded720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x5bdf1720 | 0xd20000 |
155
+ | 52 | blk.5.attn_q.weight | 0x5cb11720 | 0xb00000 |
156
+ | 53 | blk.5.attn_v.weight | 0x5d611720 | 0x348000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x5d959720 | 0x3b80000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x614d9720 | 0x2680000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x63b59720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x63b5d720 | 0x2680000 |
161
+ | 58 | blk.6.attn_k.weight | 0x661dd720 | 0x2c0000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x6649d720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x664a1720 | 0xd20000 |
164
+ | 61 | blk.6.attn_q.weight | 0x671c1720 | 0xb00000 |
165
+ | 62 | blk.6.attn_v.weight | 0x67cc1720 | 0x348000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x68009720 | 0x3b80000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x6bb89720 | 0x2680000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x6e209720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x6e20d720 | 0x2680000 |
170
+ | 67 | blk.7.attn_k.weight | 0x7088d720 | 0x2c0000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x70b4d720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x70b51720 | 0xd20000 |
173
+ | 70 | blk.7.attn_q.weight | 0x71871720 | 0xb00000 |
174
+ | 71 | blk.7.attn_v.weight | 0x72371720 | 0x348000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x726b9720 | 0x3b80000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x76239720 | 0x2680000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x788b9720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x788bd720 | 0x2680000 |
179
+ | 76 | blk.8.attn_k.weight | 0x7af3d720 | 0x2c0000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x7b1fd720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x7b201720 | 0xd20000 |
182
+ | 79 | blk.8.attn_q.weight | 0x7bf21720 | 0xb00000 |
183
+ | 80 | blk.8.attn_v.weight | 0x7ca21720 | 0x348000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x7cd69720 | 0x3b80000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x808e9720 | 0x2680000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x82f69720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x82f6d720 | 0x2680000 |
188
+ | 85 | blk.9.attn_k.weight | 0x855ed720 | 0x2c0000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x858ad720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x858b1720 | 0xd20000 |
191
+ | 88 | blk.9.attn_q.weight | 0x865d1720 | 0xb00000 |
192
+ | 89 | blk.9.attn_v.weight | 0x870d1720 | 0x348000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x87419720 | 0x3b80000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x8af99720 | 0x2680000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0x8d619720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0x8d61d720 | 0x2680000 |
197
+ | 94 | blk.10.attn_k.weight | 0x8fc9d720 | 0x2c0000 |
198
+ | 95 | blk.10.attn_norm.weight | 0x8ff5d720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0x8ff61720 | 0xd20000 |
200
+ | 97 | blk.10.attn_q.weight | 0x90c81720 | 0xb00000 |
201
+ | 98 | blk.10.attn_v.weight | 0x91781720 | 0x348000 |
202
+ | 99 | blk.10.ffn_down.weight | 0x91ac9720 | 0x3b80000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0x95649720 | 0x2680000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0x97cc9720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0x97ccd720 | 0x2680000 |
206
+ | 103 | blk.11.attn_k.weight | 0x9a34d720 | 0x2c0000 |
207
+ | 104 | blk.11.attn_norm.weight | 0x9a60d720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0x9a611720 | 0xd20000 |
209
+ | 106 | blk.11.attn_q.weight | 0x9b331720 | 0xb00000 |
210
+ | 107 | blk.11.attn_v.weight | 0x9be31720 | 0x348000 |
211
+ | 108 | blk.11.ffn_down.weight | 0x9c179720 | 0x3b80000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0x9fcf9720 | 0x2680000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0xa2379720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0xa237d720 | 0x2680000 |
215
+ | 112 | blk.12.attn_k.weight | 0xa49fd720 | 0x2c0000 |
216
+ | 113 | blk.12.attn_norm.weight | 0xa4cbd720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0xa4cc1720 | 0xd20000 |
218
+ | 115 | blk.12.attn_q.weight | 0xa59e1720 | 0xb00000 |
219
+ | 116 | blk.12.attn_v.weight | 0xa64e1720 | 0x348000 |
220
+ | 117 | blk.12.ffn_down.weight | 0xa6829720 | 0x3b80000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0xaa3a9720 | 0x2680000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0xaca29720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0xaca2d720 | 0x2680000 |
224
+ | 121 | blk.13.attn_k.weight | 0xaf0ad720 | 0x348000 |
225
+ | 122 | blk.13.attn_norm.weight | 0xaf3f5720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0xaf3f9720 | 0xd20000 |
227
+ | 124 | blk.13.attn_q.weight | 0xb0119720 | 0xd20000 |
228
+ | 125 | blk.13.attn_v.weight | 0xb0e39720 | 0x440000 |
229
+ | 126 | blk.13.ffn_down.weight | 0xb1279720 | 0x3b80000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0xb4df9720 | 0x2680000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0xb7479720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0xb747d720 | 0x2680000 |
233
+ | 130 | blk.14.attn_k.weight | 0xb9afd720 | 0x348000 |
234
+ | 131 | blk.14.attn_norm.weight | 0xb9e45720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0xb9e49720 | 0xd20000 |
236
+ | 133 | blk.14.attn_q.weight | 0xbab69720 | 0xd20000 |
237
+ | 134 | blk.14.attn_v.weight | 0xbb889720 | 0x440000 |
238
+ | 135 | blk.14.ffn_down.weight | 0xbbcc9720 | 0x3b80000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0xbf849720 | 0x2680000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0xc1ec9720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0xc1ecd720 | 0x2680000 |
242
+ | 139 | blk.15.attn_k.weight | 0xc454d720 | 0x2c0000 |
243
+ | 140 | blk.15.attn_norm.weight | 0xc480d720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0xc4811720 | 0xd20000 |
245
+ | 142 | blk.15.attn_q.weight | 0xc5531720 | 0xb00000 |
246
+ | 143 | blk.15.attn_v.weight | 0xc6031720 | 0x348000 |
247
+ | 144 | blk.15.ffn_down.weight | 0xc6379720 | 0x3b80000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0xc9ef9720 | 0x2680000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0xcc579720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0xcc57d720 | 0x2680000 |
251
+ | 148 | blk.16.attn_k.weight | 0xcebfd720 | 0x348000 |
252
+ | 149 | blk.16.attn_norm.weight | 0xcef45720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0xcef49720 | 0xd20000 |
254
+ | 151 | blk.16.attn_q.weight | 0xcfc69720 | 0xd20000 |
255
+ | 152 | blk.16.attn_v.weight | 0xd0989720 | 0x440000 |
256
+ | 153 | blk.16.ffn_down.weight | 0xd0dc9720 | 0x3b80000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0xd4949720 | 0x2df0000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0xd7739720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0xd773d720 | 0x2df0000 |
260
+ | 157 | blk.17.attn_k.weight | 0xda52d720 | 0x2c0000 |
261
+ | 158 | blk.17.attn_norm.weight | 0xda7ed720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0xda7f1720 | 0xd20000 |
263
+ | 160 | blk.17.attn_q.weight | 0xdb511720 | 0xb00000 |
264
+ | 161 | blk.17.attn_v.weight | 0xdc011720 | 0x348000 |
265
+ | 162 | blk.17.ffn_down.weight | 0xdc359720 | 0x3b80000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0xdfed9720 | 0x2df0000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0xe2cc9720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0xe2ccd720 | 0x2df0000 |
269
+ | 166 | blk.18.attn_k.weight | 0xe5abd720 | 0x348000 |
270
+ | 167 | blk.18.attn_norm.weight | 0xe5e05720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0xe5e09720 | 0xd20000 |
272
+ | 169 | blk.18.attn_q.weight | 0xe6b29720 | 0xd20000 |
273
+ | 170 | blk.18.attn_v.weight | 0xe7849720 | 0x440000 |
274
+ | 171 | blk.18.ffn_down.weight | 0xe7c89720 | 0x3b80000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0xeb809720 | 0x2df0000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0xee5f9720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0xee5fd720 | 0x2df0000 |
278
+ | 175 | blk.19.attn_k.weight | 0xf13ed720 | 0x348000 |
279
+ | 176 | blk.19.attn_norm.weight | 0xf1735720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0xf1739720 | 0xd20000 |
281
+ | 178 | blk.19.attn_q.weight | 0xf2459720 | 0xd20000 |
282
+ | 179 | blk.19.attn_v.weight | 0xf3179720 | 0x440000 |
283
+ | 180 | blk.19.ffn_down.weight | 0xf35b9720 | 0x3b80000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0xf7139720 | 0x2df0000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0xf9f29720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0xf9f2d720 | 0x2df0000 |
287
+ | 184 | blk.20.attn_k.weight | 0xfcd1d720 | 0x348000 |
288
+ | 185 | blk.20.attn_norm.weight | 0xfd065720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0xfd069720 | 0xd20000 |
290
+ | 187 | blk.20.attn_q.weight | 0xfdd89720 | 0xd20000 |
291
+ | 188 | blk.20.attn_v.weight | 0xfeaa9720 | 0x440000 |
292
+ | 189 | blk.20.ffn_down.weight | 0xfeee9720 | 0x3b80000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0x102a69720 | 0x2df0000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0x105859720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0x10585d720 | 0x2df0000 |
296
+ | 193 | blk.21.attn_k.weight | 0x10864d720 | 0x348000 |
297
+ | 194 | blk.21.attn_norm.weight | 0x108995720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0x108999720 | 0xd20000 |
299
+ | 196 | blk.21.attn_q.weight | 0x1096b9720 | 0xd20000 |
300
+ | 197 | blk.21.attn_v.weight | 0x10a3d9720 | 0x440000 |
301
+ | 198 | blk.21.ffn_down.weight | 0x10a819720 | 0x3b80000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0x10e399720 | 0x2df0000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0x111189720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0x11118d720 | 0x2df0000 |
305
+ | 202 | blk.22.attn_k.weight | 0x113f7d720 | 0x348000 |
306
+ | 203 | blk.22.attn_norm.weight | 0x1142c5720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0x1142c9720 | 0xd20000 |
308
+ | 205 | blk.22.attn_q.weight | 0x114fe9720 | 0xd20000 |
309
+ | 206 | blk.22.attn_v.weight | 0x115d09720 | 0x440000 |
310
+ | 207 | blk.22.ffn_down.weight | 0x116149720 | 0x3b80000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0x119cc9720 | 0x2df0000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0x11cab9720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0x11cabd720 | 0x2df0000 |
314
+ | 211 | blk.23.attn_k.weight | 0x11f8ad720 | 0x348000 |
315
+ | 212 | blk.23.attn_norm.weight | 0x11fbf5720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0x11fbf9720 | 0xd20000 |
317
+ | 214 | blk.23.attn_q.weight | 0x120919720 | 0xd20000 |
318
+ | 215 | blk.23.attn_v.weight | 0x121639720 | 0x440000 |
319
+ | 216 | blk.23.ffn_down.weight | 0x121a79720 | 0x3b80000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0x1255f9720 | 0x2df0000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0x1283e9720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0x1283ed720 | 0x2df0000 |
323
+ | 220 | blk.24.attn_k.weight | 0x12b1dd720 | 0x348000 |
324
+ | 221 | blk.24.attn_norm.weight | 0x12b525720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0x12b529720 | 0xd20000 |
326
+ | 223 | blk.24.attn_q.weight | 0x12c249720 | 0xd20000 |
327
+ | 224 | blk.24.attn_v.weight | 0x12cf69720 | 0x440000 |
328
+ | 225 | blk.24.ffn_down.weight | 0x12d3a9720 | 0x3b80000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0x130f29720 | 0x2df0000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0x133d19720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0x133d1d720 | 0x2df0000 |
332
+ | 229 | blk.25.attn_k.weight | 0x136b0d720 | 0x348000 |
333
+ | 230 | blk.25.attn_norm.weight | 0x136e55720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0x136e59720 | 0xd20000 |
335
+ | 232 | blk.25.attn_q.weight | 0x137b79720 | 0xd20000 |
336
+ | 233 | blk.25.attn_v.weight | 0x138899720 | 0x440000 |
337
+ | 234 | blk.25.ffn_down.weight | 0x138cd9720 | 0x3b80000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0x13c859720 | 0x2df0000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0x13f649720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0x13f64d720 | 0x2df0000 |
341
+ | 238 | blk.26.attn_k.weight | 0x14243d720 | 0x348000 |
342
+ | 239 | blk.26.attn_norm.weight | 0x142785720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0x142789720 | 0xd20000 |
344
+ | 241 | blk.26.attn_q.weight | 0x1434a9720 | 0xd20000 |
345
+ | 242 | blk.26.attn_v.weight | 0x1441c9720 | 0x440000 |
346
+ | 243 | blk.26.ffn_down.weight | 0x144609720 | 0x3b80000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0x148189720 | 0x2df0000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0x14af79720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0x14af7d720 | 0x2df0000 |
350
+ | 247 | blk.27.attn_k.weight | 0x14dd6d720 | 0x348000 |
351
+ | 248 | blk.27.attn_norm.weight | 0x14e0b5720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0x14e0b9720 | 0xd20000 |
353
+ | 250 | blk.27.attn_q.weight | 0x14edd9720 | 0xd20000 |
354
+ | 251 | blk.27.attn_v.weight | 0x14faf9720 | 0x440000 |
355
+ | 252 | blk.27.ffn_down.weight | 0x14ff39720 | 0x3b80000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0x153ab9720 | 0x2df0000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0x1568a9720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0x1568ad720 | 0x2df0000 |
359
+ | 256 | blk.28.attn_k.weight | 0x15969d720 | 0x348000 |
360
+ | 257 | blk.28.attn_norm.weight | 0x1599e5720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0x1599e9720 | 0xd20000 |
362
+ | 259 | blk.28.attn_q.weight | 0x15a709720 | 0xd20000 |
363
+ | 260 | blk.28.attn_v.weight | 0x15b429720 | 0x440000 |
364
+ | 261 | blk.28.ffn_down.weight | 0x15b869720 | 0x3b80000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0x15f3e9720 | 0x2df0000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0x1621d9720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0x1621dd720 | 0x2df0000 |
368
+ | 265 | blk.29.attn_k.weight | 0x164fcd720 | 0x348000 |
369
+ | 266 | blk.29.attn_norm.weight | 0x165315720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0x165319720 | 0xd20000 |
371
+ | 268 | blk.29.attn_q.weight | 0x166039720 | 0xd20000 |
372
+ | 269 | blk.29.attn_v.weight | 0x166d59720 | 0x440000 |
373
+ | 270 | blk.29.ffn_down.weight | 0x167199720 | 0x3b80000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0x16ad19720 | 0x2df0000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0x16db09720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0x16db0d720 | 0x2df0000 |
377
+ | 274 | blk.30.attn_k.weight | 0x1708fd720 | 0x348000 |
378
+ | 275 | blk.30.attn_norm.weight | 0x170c45720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0x170c49720 | 0xd20000 |
380
+ | 277 | blk.30.attn_q.weight | 0x171969720 | 0xd20000 |
381
+ | 278 | blk.30.attn_v.weight | 0x172689720 | 0x440000 |
382
+ | 279 | blk.30.ffn_down.weight | 0x172ac9720 | 0x3b80000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0x176649720 | 0x2df0000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0x179439720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0x17943d720 | 0x2df0000 |
386
+ | 283 | blk.31.attn_k.weight | 0x17c22d720 | 0x2c0000 |
387
+ | 284 | blk.31.attn_norm.weight | 0x17c4ed720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0x17c4f1720 | 0xd20000 |
389
+ | 286 | blk.31.attn_q.weight | 0x17d211720 | 0xb00000 |
390
+ | 287 | blk.31.attn_v.weight | 0x17dd11720 | 0x348000 |
391
+ | 288 | blk.31.ffn_down.weight | 0x17e059720 | 0x3b80000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0x181bd9720 | 0x2df0000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0x1849c9720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0x1849cd720 | 0x2df0000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q6_K |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q3_K |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q5_K |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q5_K |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q5_K |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%
scores/Watt-Tool-8B-Q8_0.md ADDED
@@ -0,0 +1,982 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Watt-Tool-8B-Q8_0.gguf - GGUF Internal File Dump
2
+
3
+ - Endian: LITTLE endian
4
+
5
+ ## Key Value Metadata Store
6
+
7
+ There are 43 key-value pairs in this file
8
+
9
+ | POS | TYPE | Count | Key | Value |
10
+ |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------|
11
+ | 1 | UINT32 | 1 | GGUF.version | 3 |
12
+ | 2 | UINT64 | 1 | GGUF.tensor_count | 292 |
13
+ | 3 | UINT64 | 1 | GGUF.kv_count | 40 |
14
+ | 4 | STRING | 1 | general.architecture | `llama` |
15
+ | 5 | STRING | 1 | general.type | `model` |
16
+ | 6 | STRING | 1 | general.name | `Watt Tool 8B GGUF` |
17
+ | 7 | STRING | 1 | general.finetune | `GGUF` |
18
+ | 8 | STRING | 1 | general.basename | `Watt-Tool` |
19
+ | 9 | STRING | 1 | general.size_label | `8B` |
20
+ | 10 | STRING | 1 | general.license | `apache-2.0` |
21
+ | 11 | UINT32 | 1 | general.base_model.count | 1 |
22
+ | 12 | STRING | 1 | general.base_model.0.name | `Llama 3.1 8B Instruct` |
23
+ | 13 | STRING | 1 | general.base_model.0.organization | `Meta Llama` |
24
+ | 14 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct` |
25
+ | 15 | [STRING] | 4 | general.tags | [ `function-calling`, `tool-use`, `llama`, `bfcl` ] |
26
+ | 16 | [STRING] | 1 | general.languages | [ `en` ] |
27
+ | 17 | UINT32 | 1 | llama.block_count | 32 |
28
+ | 18 | UINT32 | 1 | llama.context_length | 131072 |
29
+ | 19 | UINT32 | 1 | llama.embedding_length | 4096 |
30
+ | 20 | UINT32 | 1 | llama.feed_forward_length | 14336 |
31
+ | 21 | UINT32 | 1 | llama.attention.head_count | 32 |
32
+ | 22 | UINT32 | 1 | llama.attention.head_count_kv | 8 |
33
+ | 23 | FLOAT32 | 1 | llama.rope.freq_base | 500000.0 |
34
+ | 24 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 |
35
+ | 25 | UINT32 | 1 | llama.attention.key_length | 128 |
36
+ | 26 | UINT32 | 1 | llama.attention.value_length | 128 |
37
+ | 27 | UINT32 | 1 | llama.vocab_size | 128256 |
38
+ | 28 | UINT32 | 1 | llama.rope.dimension_count | 128 |
39
+ | 29 | STRING | 1 | tokenizer.ggml.model | `gpt2` |
40
+ | 30 | STRING | 1 | tokenizer.ggml.pre | `llama-bpe` |
41
+ | 31 | [STRING] | 128256 | tokenizer.ggml.tokens | [ `!`, `"`, `#`, `$`, `%`, ... ] |
42
+ | 32 | [INT32] | 128256 | tokenizer.ggml.token_type | [ 1, 1, 1, 1, 1, 1, 1, ... ] |
43
+ | 33 | [STRING] | 280147 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ ĠĠĠ`, `ĠĠ ĠĠ`, `ĠĠĠ Ġ`, `i n`, ... ] |
44
+ | 34 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 128000 |
45
+ | 35 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 128009 |
46
+ | 36 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 128009 |
47
+ | 37 | STRING | 1 | tokenizer.chat_template | `{{ '<|begin_of_text|>' }}{% if`...`d|>' }}{% endif %}{% endfor %}` |
48
+ | 38 | UINT32 | 1 | general.quantization_version | 2 |
49
+ | 39 | UINT32 | 1 | general.file_type | 7 |
50
+ | 40 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Watt-Tool-8B-small.dat` |
51
+ | 41 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/calibration_eur_small.txt` |
52
+ | 42 | INT32 | 1 | quantize.imatrix.entries_count | 225 |
53
+ | 43 | INT32 | 1 | quantize.imatrix.chunks_count | 962 |
54
+
55
+ ## Tensors Overview ~8B Elements
56
+
57
+ Total number of elements in all tensors: 8030261312 Elements
58
+
59
+ - [Watt-Tool-8B-Q8\_0.gguf - GGUF Internal File Dump](#watt-tool-8b-q8_0gguf---gguf-internal-file-dump)
60
+ - [Key Value Metadata Store](#key-value-metadata-store)
61
+ - [Tensors Overview ~8B Elements](#tensors-overview-8b-elements)
62
+ - [Tensor Data Offset](#tensor-data-offset)
63
+ - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements)
64
+ - [Block 0 Tensor Group : ~218M Elements](#block-0-tensor-group--218m-elements)
65
+ - [Block 1 Tensor Group : ~218M Elements](#block-1-tensor-group--218m-elements)
66
+ - [Block 2 Tensor Group : ~218M Elements](#block-2-tensor-group--218m-elements)
67
+ - [Block 3 Tensor Group : ~218M Elements](#block-3-tensor-group--218m-elements)
68
+ - [Block 4 Tensor Group : ~218M Elements](#block-4-tensor-group--218m-elements)
69
+ - [Block 5 Tensor Group : ~218M Elements](#block-5-tensor-group--218m-elements)
70
+ - [Block 6 Tensor Group : ~218M Elements](#block-6-tensor-group--218m-elements)
71
+ - [Block 7 Tensor Group : ~218M Elements](#block-7-tensor-group--218m-elements)
72
+ - [Block 8 Tensor Group : ~218M Elements](#block-8-tensor-group--218m-elements)
73
+ - [Block 9 Tensor Group : ~218M Elements](#block-9-tensor-group--218m-elements)
74
+ - [Block 10 Tensor Group : ~218M Elements](#block-10-tensor-group--218m-elements)
75
+ - [Block 11 Tensor Group : ~218M Elements](#block-11-tensor-group--218m-elements)
76
+ - [Block 12 Tensor Group : ~218M Elements](#block-12-tensor-group--218m-elements)
77
+ - [Block 13 Tensor Group : ~218M Elements](#block-13-tensor-group--218m-elements)
78
+ - [Block 14 Tensor Group : ~218M Elements](#block-14-tensor-group--218m-elements)
79
+ - [Block 15 Tensor Group : ~218M Elements](#block-15-tensor-group--218m-elements)
80
+ - [Block 16 Tensor Group : ~218M Elements](#block-16-tensor-group--218m-elements)
81
+ - [Block 17 Tensor Group : ~218M Elements](#block-17-tensor-group--218m-elements)
82
+ - [Block 18 Tensor Group : ~218M Elements](#block-18-tensor-group--218m-elements)
83
+ - [Block 19 Tensor Group : ~218M Elements](#block-19-tensor-group--218m-elements)
84
+ - [Block 20 Tensor Group : ~218M Elements](#block-20-tensor-group--218m-elements)
85
+ - [Block 21 Tensor Group : ~218M Elements](#block-21-tensor-group--218m-elements)
86
+ - [Block 22 Tensor Group : ~218M Elements](#block-22-tensor-group--218m-elements)
87
+ - [Block 23 Tensor Group : ~218M Elements](#block-23-tensor-group--218m-elements)
88
+ - [Block 24 Tensor Group : ~218M Elements](#block-24-tensor-group--218m-elements)
89
+ - [Block 25 Tensor Group : ~218M Elements](#block-25-tensor-group--218m-elements)
90
+ - [Block 26 Tensor Group : ~218M Elements](#block-26-tensor-group--218m-elements)
91
+ - [Block 27 Tensor Group : ~218M Elements](#block-27-tensor-group--218m-elements)
92
+ - [Block 28 Tensor Group : ~218M Elements](#block-28-tensor-group--218m-elements)
93
+ - [Block 29 Tensor Group : ~218M Elements](#block-29-tensor-group--218m-elements)
94
+ - [Block 30 Tensor Group : ~218M Elements](#block-30-tensor-group--218m-elements)
95
+ - [Block 31 Tensor Group : ~218M Elements](#block-31-tensor-group--218m-elements)
96
+
97
+ ### Tensor Data Offset
98
+
99
+ This table contains the offset and data segment relative to start of file
100
+
101
+ | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) |
102
+ |-----:|:--------------------------|-----------------:|-----------------:|
103
+ | 0 | output.weight | 0x779620 | 0x21450000 |
104
+ | 1 | output_norm.weight | 0x21bc9620 | 0x4000 |
105
+ | 2 | rope_freqs.weight | 0x21bcd620 | 0x100 |
106
+ | 3 | token_embd.weight | 0x21bcd720 | 0xd746000 |
107
+ | 4 | blk.0.attn_k.weight | 0x2f313720 | 0x348000 |
108
+ | 5 | blk.0.attn_norm.weight | 0x2f65b720 | 0x4000 |
109
+ | 6 | blk.0.attn_output.weight | 0x2f65f720 | 0x1100000 |
110
+ | 7 | blk.0.attn_q.weight | 0x3075f720 | 0xd20000 |
111
+ | 8 | blk.0.attn_v.weight | 0x3147f720 | 0x440000 |
112
+ | 9 | blk.0.ffn_down.weight | 0x318bf720 | 0x3b80000 |
113
+ | 10 | blk.0.ffn_gate.weight | 0x3543f720 | 0x2df0000 |
114
+ | 11 | blk.0.ffn_norm.weight | 0x3822f720 | 0x4000 |
115
+ | 12 | blk.0.ffn_up.weight | 0x38233720 | 0x2df0000 |
116
+ | 13 | blk.1.attn_k.weight | 0x3b023720 | 0x348000 |
117
+ | 14 | blk.1.attn_norm.weight | 0x3b36b720 | 0x4000 |
118
+ | 15 | blk.1.attn_output.weight | 0x3b36f720 | 0x1100000 |
119
+ | 16 | blk.1.attn_q.weight | 0x3c46f720 | 0xd20000 |
120
+ | 17 | blk.1.attn_v.weight | 0x3d18f720 | 0x440000 |
121
+ | 18 | blk.1.ffn_down.weight | 0x3d5cf720 | 0x3b80000 |
122
+ | 19 | blk.1.ffn_gate.weight | 0x4114f720 | 0x2df0000 |
123
+ | 20 | blk.1.ffn_norm.weight | 0x43f3f720 | 0x4000 |
124
+ | 21 | blk.1.ffn_up.weight | 0x43f43720 | 0x2df0000 |
125
+ | 22 | blk.2.attn_k.weight | 0x46d33720 | 0x348000 |
126
+ | 23 | blk.2.attn_norm.weight | 0x4707b720 | 0x4000 |
127
+ | 24 | blk.2.attn_output.weight | 0x4707f720 | 0x1100000 |
128
+ | 25 | blk.2.attn_q.weight | 0x4817f720 | 0xd20000 |
129
+ | 26 | blk.2.attn_v.weight | 0x48e9f720 | 0x440000 |
130
+ | 27 | blk.2.ffn_down.weight | 0x492df720 | 0x3b80000 |
131
+ | 28 | blk.2.ffn_gate.weight | 0x4ce5f720 | 0x2df0000 |
132
+ | 29 | blk.2.ffn_norm.weight | 0x4fc4f720 | 0x4000 |
133
+ | 30 | blk.2.ffn_up.weight | 0x4fc53720 | 0x2df0000 |
134
+ | 31 | blk.3.attn_k.weight | 0x52a43720 | 0x348000 |
135
+ | 32 | blk.3.attn_norm.weight | 0x52d8b720 | 0x4000 |
136
+ | 33 | blk.3.attn_output.weight | 0x52d8f720 | 0x1100000 |
137
+ | 34 | blk.3.attn_q.weight | 0x53e8f720 | 0xd20000 |
138
+ | 35 | blk.3.attn_v.weight | 0x54baf720 | 0x440000 |
139
+ | 36 | blk.3.ffn_down.weight | 0x54fef720 | 0x3b80000 |
140
+ | 37 | blk.3.ffn_gate.weight | 0x58b6f720 | 0x2df0000 |
141
+ | 38 | blk.3.ffn_norm.weight | 0x5b95f720 | 0x4000 |
142
+ | 39 | blk.3.ffn_up.weight | 0x5b963720 | 0x2df0000 |
143
+ | 40 | blk.4.attn_k.weight | 0x5e753720 | 0x348000 |
144
+ | 41 | blk.4.attn_norm.weight | 0x5ea9b720 | 0x4000 |
145
+ | 42 | blk.4.attn_output.weight | 0x5ea9f720 | 0x1100000 |
146
+ | 43 | blk.4.attn_q.weight | 0x5fb9f720 | 0xd20000 |
147
+ | 44 | blk.4.attn_v.weight | 0x608bf720 | 0x440000 |
148
+ | 45 | blk.4.ffn_down.weight | 0x60cff720 | 0x3b80000 |
149
+ | 46 | blk.4.ffn_gate.weight | 0x6487f720 | 0x2df0000 |
150
+ | 47 | blk.4.ffn_norm.weight | 0x6766f720 | 0x4000 |
151
+ | 48 | blk.4.ffn_up.weight | 0x67673720 | 0x2df0000 |
152
+ | 49 | blk.5.attn_k.weight | 0x6a463720 | 0x348000 |
153
+ | 50 | blk.5.attn_norm.weight | 0x6a7ab720 | 0x4000 |
154
+ | 51 | blk.5.attn_output.weight | 0x6a7af720 | 0x1100000 |
155
+ | 52 | blk.5.attn_q.weight | 0x6b8af720 | 0xd20000 |
156
+ | 53 | blk.5.attn_v.weight | 0x6c5cf720 | 0x440000 |
157
+ | 54 | blk.5.ffn_down.weight | 0x6ca0f720 | 0x3b80000 |
158
+ | 55 | blk.5.ffn_gate.weight | 0x7058f720 | 0x2df0000 |
159
+ | 56 | blk.5.ffn_norm.weight | 0x7337f720 | 0x4000 |
160
+ | 57 | blk.5.ffn_up.weight | 0x73383720 | 0x2df0000 |
161
+ | 58 | blk.6.attn_k.weight | 0x76173720 | 0x348000 |
162
+ | 59 | blk.6.attn_norm.weight | 0x764bb720 | 0x4000 |
163
+ | 60 | blk.6.attn_output.weight | 0x764bf720 | 0x1100000 |
164
+ | 61 | blk.6.attn_q.weight | 0x775bf720 | 0xd20000 |
165
+ | 62 | blk.6.attn_v.weight | 0x782df720 | 0x440000 |
166
+ | 63 | blk.6.ffn_down.weight | 0x7871f720 | 0x3b80000 |
167
+ | 64 | blk.6.ffn_gate.weight | 0x7c29f720 | 0x2df0000 |
168
+ | 65 | blk.6.ffn_norm.weight | 0x7f08f720 | 0x4000 |
169
+ | 66 | blk.6.ffn_up.weight | 0x7f093720 | 0x2df0000 |
170
+ | 67 | blk.7.attn_k.weight | 0x81e83720 | 0x348000 |
171
+ | 68 | blk.7.attn_norm.weight | 0x821cb720 | 0x4000 |
172
+ | 69 | blk.7.attn_output.weight | 0x821cf720 | 0x1100000 |
173
+ | 70 | blk.7.attn_q.weight | 0x832cf720 | 0xd20000 |
174
+ | 71 | blk.7.attn_v.weight | 0x83fef720 | 0x440000 |
175
+ | 72 | blk.7.ffn_down.weight | 0x8442f720 | 0x3b80000 |
176
+ | 73 | blk.7.ffn_gate.weight | 0x87faf720 | 0x2df0000 |
177
+ | 74 | blk.7.ffn_norm.weight | 0x8ad9f720 | 0x4000 |
178
+ | 75 | blk.7.ffn_up.weight | 0x8ada3720 | 0x2df0000 |
179
+ | 76 | blk.8.attn_k.weight | 0x8db93720 | 0x348000 |
180
+ | 77 | blk.8.attn_norm.weight | 0x8dedb720 | 0x4000 |
181
+ | 78 | blk.8.attn_output.weight | 0x8dedf720 | 0x1100000 |
182
+ | 79 | blk.8.attn_q.weight | 0x8efdf720 | 0xd20000 |
183
+ | 80 | blk.8.attn_v.weight | 0x8fcff720 | 0x440000 |
184
+ | 81 | blk.8.ffn_down.weight | 0x9013f720 | 0x3b80000 |
185
+ | 82 | blk.8.ffn_gate.weight | 0x93cbf720 | 0x2df0000 |
186
+ | 83 | blk.8.ffn_norm.weight | 0x96aaf720 | 0x4000 |
187
+ | 84 | blk.8.ffn_up.weight | 0x96ab3720 | 0x2df0000 |
188
+ | 85 | blk.9.attn_k.weight | 0x998a3720 | 0x348000 |
189
+ | 86 | blk.9.attn_norm.weight | 0x99beb720 | 0x4000 |
190
+ | 87 | blk.9.attn_output.weight | 0x99bef720 | 0x1100000 |
191
+ | 88 | blk.9.attn_q.weight | 0x9acef720 | 0xd20000 |
192
+ | 89 | blk.9.attn_v.weight | 0x9ba0f720 | 0x440000 |
193
+ | 90 | blk.9.ffn_down.weight | 0x9be4f720 | 0x3b80000 |
194
+ | 91 | blk.9.ffn_gate.weight | 0x9f9cf720 | 0x2df0000 |
195
+ | 92 | blk.9.ffn_norm.weight | 0xa27bf720 | 0x4000 |
196
+ | 93 | blk.9.ffn_up.weight | 0xa27c3720 | 0x2df0000 |
197
+ | 94 | blk.10.attn_k.weight | 0xa55b3720 | 0x348000 |
198
+ | 95 | blk.10.attn_norm.weight | 0xa58fb720 | 0x4000 |
199
+ | 96 | blk.10.attn_output.weight | 0xa58ff720 | 0x1100000 |
200
+ | 97 | blk.10.attn_q.weight | 0xa69ff720 | 0xd20000 |
201
+ | 98 | blk.10.attn_v.weight | 0xa771f720 | 0x440000 |
202
+ | 99 | blk.10.ffn_down.weight | 0xa7b5f720 | 0x3b80000 |
203
+ | 100 | blk.10.ffn_gate.weight | 0xab6df720 | 0x2df0000 |
204
+ | 101 | blk.10.ffn_norm.weight | 0xae4cf720 | 0x4000 |
205
+ | 102 | blk.10.ffn_up.weight | 0xae4d3720 | 0x2df0000 |
206
+ | 103 | blk.11.attn_k.weight | 0xb12c3720 | 0x348000 |
207
+ | 104 | blk.11.attn_norm.weight | 0xb160b720 | 0x4000 |
208
+ | 105 | blk.11.attn_output.weight | 0xb160f720 | 0x1100000 |
209
+ | 106 | blk.11.attn_q.weight | 0xb270f720 | 0xd20000 |
210
+ | 107 | blk.11.attn_v.weight | 0xb342f720 | 0x440000 |
211
+ | 108 | blk.11.ffn_down.weight | 0xb386f720 | 0x3b80000 |
212
+ | 109 | blk.11.ffn_gate.weight | 0xb73ef720 | 0x2df0000 |
213
+ | 110 | blk.11.ffn_norm.weight | 0xba1df720 | 0x4000 |
214
+ | 111 | blk.11.ffn_up.weight | 0xba1e3720 | 0x2df0000 |
215
+ | 112 | blk.12.attn_k.weight | 0xbcfd3720 | 0x348000 |
216
+ | 113 | blk.12.attn_norm.weight | 0xbd31b720 | 0x4000 |
217
+ | 114 | blk.12.attn_output.weight | 0xbd31f720 | 0x1100000 |
218
+ | 115 | blk.12.attn_q.weight | 0xbe41f720 | 0xd20000 |
219
+ | 116 | blk.12.attn_v.weight | 0xbf13f720 | 0x440000 |
220
+ | 117 | blk.12.ffn_down.weight | 0xbf57f720 | 0x3b80000 |
221
+ | 118 | blk.12.ffn_gate.weight | 0xc30ff720 | 0x2df0000 |
222
+ | 119 | blk.12.ffn_norm.weight | 0xc5eef720 | 0x4000 |
223
+ | 120 | blk.12.ffn_up.weight | 0xc5ef3720 | 0x2df0000 |
224
+ | 121 | blk.13.attn_k.weight | 0xc8ce3720 | 0x440000 |
225
+ | 122 | blk.13.attn_norm.weight | 0xc9123720 | 0x4000 |
226
+ | 123 | blk.13.attn_output.weight | 0xc9127720 | 0x1100000 |
227
+ | 124 | blk.13.attn_q.weight | 0xca227720 | 0x1100000 |
228
+ | 125 | blk.13.attn_v.weight | 0xcb327720 | 0x800000 |
229
+ | 126 | blk.13.ffn_down.weight | 0xcbb27720 | 0x3b80000 |
230
+ | 127 | blk.13.ffn_gate.weight | 0xcf6a7720 | 0x2df0000 |
231
+ | 128 | blk.13.ffn_norm.weight | 0xd2497720 | 0x4000 |
232
+ | 129 | blk.13.ffn_up.weight | 0xd249b720 | 0x2df0000 |
233
+ | 130 | blk.14.attn_k.weight | 0xd528b720 | 0x440000 |
234
+ | 131 | blk.14.attn_norm.weight | 0xd56cb720 | 0x4000 |
235
+ | 132 | blk.14.attn_output.weight | 0xd56cf720 | 0x1100000 |
236
+ | 133 | blk.14.attn_q.weight | 0xd67cf720 | 0x1100000 |
237
+ | 134 | blk.14.attn_v.weight | 0xd78cf720 | 0x800000 |
238
+ | 135 | blk.14.ffn_down.weight | 0xd80cf720 | 0x3b80000 |
239
+ | 136 | blk.14.ffn_gate.weight | 0xdbc4f720 | 0x2df0000 |
240
+ | 137 | blk.14.ffn_norm.weight | 0xdea3f720 | 0x4000 |
241
+ | 138 | blk.14.ffn_up.weight | 0xdea43720 | 0x2df0000 |
242
+ | 139 | blk.15.attn_k.weight | 0xe1833720 | 0x348000 |
243
+ | 140 | blk.15.attn_norm.weight | 0xe1b7b720 | 0x4000 |
244
+ | 141 | blk.15.attn_output.weight | 0xe1b7f720 | 0x1100000 |
245
+ | 142 | blk.15.attn_q.weight | 0xe2c7f720 | 0xd20000 |
246
+ | 143 | blk.15.attn_v.weight | 0xe399f720 | 0x440000 |
247
+ | 144 | blk.15.ffn_down.weight | 0xe3ddf720 | 0x3b80000 |
248
+ | 145 | blk.15.ffn_gate.weight | 0xe795f720 | 0x2df0000 |
249
+ | 146 | blk.15.ffn_norm.weight | 0xea74f720 | 0x4000 |
250
+ | 147 | blk.15.ffn_up.weight | 0xea753720 | 0x2df0000 |
251
+ | 148 | blk.16.attn_k.weight | 0xed543720 | 0x440000 |
252
+ | 149 | blk.16.attn_norm.weight | 0xed983720 | 0x4000 |
253
+ | 150 | blk.16.attn_output.weight | 0xed987720 | 0x1100000 |
254
+ | 151 | blk.16.attn_q.weight | 0xeea87720 | 0x1100000 |
255
+ | 152 | blk.16.attn_v.weight | 0xefb87720 | 0x800000 |
256
+ | 153 | blk.16.ffn_down.weight | 0xf0387720 | 0x3b80000 |
257
+ | 154 | blk.16.ffn_gate.weight | 0xf3f07720 | 0x3b80000 |
258
+ | 155 | blk.16.ffn_norm.weight | 0xf7a87720 | 0x4000 |
259
+ | 156 | blk.16.ffn_up.weight | 0xf7a8b720 | 0x3b80000 |
260
+ | 157 | blk.17.attn_k.weight | 0xfb60b720 | 0x348000 |
261
+ | 158 | blk.17.attn_norm.weight | 0xfb953720 | 0x4000 |
262
+ | 159 | blk.17.attn_output.weight | 0xfb957720 | 0x1100000 |
263
+ | 160 | blk.17.attn_q.weight | 0xfca57720 | 0xd20000 |
264
+ | 161 | blk.17.attn_v.weight | 0xfd777720 | 0x440000 |
265
+ | 162 | blk.17.ffn_down.weight | 0xfdbb7720 | 0x3b80000 |
266
+ | 163 | blk.17.ffn_gate.weight | 0x101737720 | 0x3b80000 |
267
+ | 164 | blk.17.ffn_norm.weight | 0x1052b7720 | 0x4000 |
268
+ | 165 | blk.17.ffn_up.weight | 0x1052bb720 | 0x3b80000 |
269
+ | 166 | blk.18.attn_k.weight | 0x108e3b720 | 0x440000 |
270
+ | 167 | blk.18.attn_norm.weight | 0x10927b720 | 0x4000 |
271
+ | 168 | blk.18.attn_output.weight | 0x10927f720 | 0x1100000 |
272
+ | 169 | blk.18.attn_q.weight | 0x10a37f720 | 0x1100000 |
273
+ | 170 | blk.18.attn_v.weight | 0x10b47f720 | 0x800000 |
274
+ | 171 | blk.18.ffn_down.weight | 0x10bc7f720 | 0x3b80000 |
275
+ | 172 | blk.18.ffn_gate.weight | 0x10f7ff720 | 0x3b80000 |
276
+ | 173 | blk.18.ffn_norm.weight | 0x11337f720 | 0x4000 |
277
+ | 174 | blk.18.ffn_up.weight | 0x113383720 | 0x3b80000 |
278
+ | 175 | blk.19.attn_k.weight | 0x116f03720 | 0x440000 |
279
+ | 176 | blk.19.attn_norm.weight | 0x117343720 | 0x4000 |
280
+ | 177 | blk.19.attn_output.weight | 0x117347720 | 0x1100000 |
281
+ | 178 | blk.19.attn_q.weight | 0x118447720 | 0x1100000 |
282
+ | 179 | blk.19.attn_v.weight | 0x119547720 | 0x800000 |
283
+ | 180 | blk.19.ffn_down.weight | 0x119d47720 | 0x3b80000 |
284
+ | 181 | blk.19.ffn_gate.weight | 0x11d8c7720 | 0x3b80000 |
285
+ | 182 | blk.19.ffn_norm.weight | 0x121447720 | 0x4000 |
286
+ | 183 | blk.19.ffn_up.weight | 0x12144b720 | 0x3b80000 |
287
+ | 184 | blk.20.attn_k.weight | 0x124fcb720 | 0x440000 |
288
+ | 185 | blk.20.attn_norm.weight | 0x12540b720 | 0x4000 |
289
+ | 186 | blk.20.attn_output.weight | 0x12540f720 | 0x1100000 |
290
+ | 187 | blk.20.attn_q.weight | 0x12650f720 | 0x1100000 |
291
+ | 188 | blk.20.attn_v.weight | 0x12760f720 | 0x800000 |
292
+ | 189 | blk.20.ffn_down.weight | 0x127e0f720 | 0x3b80000 |
293
+ | 190 | blk.20.ffn_gate.weight | 0x12b98f720 | 0x3b80000 |
294
+ | 191 | blk.20.ffn_norm.weight | 0x12f50f720 | 0x4000 |
295
+ | 192 | blk.20.ffn_up.weight | 0x12f513720 | 0x3b80000 |
296
+ | 193 | blk.21.attn_k.weight | 0x133093720 | 0x440000 |
297
+ | 194 | blk.21.attn_norm.weight | 0x1334d3720 | 0x4000 |
298
+ | 195 | blk.21.attn_output.weight | 0x1334d7720 | 0x1100000 |
299
+ | 196 | blk.21.attn_q.weight | 0x1345d7720 | 0x1100000 |
300
+ | 197 | blk.21.attn_v.weight | 0x1356d7720 | 0x800000 |
301
+ | 198 | blk.21.ffn_down.weight | 0x135ed7720 | 0x3b80000 |
302
+ | 199 | blk.21.ffn_gate.weight | 0x139a57720 | 0x3b80000 |
303
+ | 200 | blk.21.ffn_norm.weight | 0x13d5d7720 | 0x4000 |
304
+ | 201 | blk.21.ffn_up.weight | 0x13d5db720 | 0x3b80000 |
305
+ | 202 | blk.22.attn_k.weight | 0x14115b720 | 0x440000 |
306
+ | 203 | blk.22.attn_norm.weight | 0x14159b720 | 0x4000 |
307
+ | 204 | blk.22.attn_output.weight | 0x14159f720 | 0x1100000 |
308
+ | 205 | blk.22.attn_q.weight | 0x14269f720 | 0x1100000 |
309
+ | 206 | blk.22.attn_v.weight | 0x14379f720 | 0x800000 |
310
+ | 207 | blk.22.ffn_down.weight | 0x143f9f720 | 0x3b80000 |
311
+ | 208 | blk.22.ffn_gate.weight | 0x147b1f720 | 0x3b80000 |
312
+ | 209 | blk.22.ffn_norm.weight | 0x14b69f720 | 0x4000 |
313
+ | 210 | blk.22.ffn_up.weight | 0x14b6a3720 | 0x3b80000 |
314
+ | 211 | blk.23.attn_k.weight | 0x14f223720 | 0x440000 |
315
+ | 212 | blk.23.attn_norm.weight | 0x14f663720 | 0x4000 |
316
+ | 213 | blk.23.attn_output.weight | 0x14f667720 | 0x1100000 |
317
+ | 214 | blk.23.attn_q.weight | 0x150767720 | 0x1100000 |
318
+ | 215 | blk.23.attn_v.weight | 0x151867720 | 0x800000 |
319
+ | 216 | blk.23.ffn_down.weight | 0x152067720 | 0x3b80000 |
320
+ | 217 | blk.23.ffn_gate.weight | 0x155be7720 | 0x3b80000 |
321
+ | 218 | blk.23.ffn_norm.weight | 0x159767720 | 0x4000 |
322
+ | 219 | blk.23.ffn_up.weight | 0x15976b720 | 0x3b80000 |
323
+ | 220 | blk.24.attn_k.weight | 0x15d2eb720 | 0x440000 |
324
+ | 221 | blk.24.attn_norm.weight | 0x15d72b720 | 0x4000 |
325
+ | 222 | blk.24.attn_output.weight | 0x15d72f720 | 0x1100000 |
326
+ | 223 | blk.24.attn_q.weight | 0x15e82f720 | 0x1100000 |
327
+ | 224 | blk.24.attn_v.weight | 0x15f92f720 | 0x800000 |
328
+ | 225 | blk.24.ffn_down.weight | 0x16012f720 | 0x3b80000 |
329
+ | 226 | blk.24.ffn_gate.weight | 0x163caf720 | 0x3b80000 |
330
+ | 227 | blk.24.ffn_norm.weight | 0x16782f720 | 0x4000 |
331
+ | 228 | blk.24.ffn_up.weight | 0x167833720 | 0x3b80000 |
332
+ | 229 | blk.25.attn_k.weight | 0x16b3b3720 | 0x440000 |
333
+ | 230 | blk.25.attn_norm.weight | 0x16b7f3720 | 0x4000 |
334
+ | 231 | blk.25.attn_output.weight | 0x16b7f7720 | 0x1100000 |
335
+ | 232 | blk.25.attn_q.weight | 0x16c8f7720 | 0x1100000 |
336
+ | 233 | blk.25.attn_v.weight | 0x16d9f7720 | 0x800000 |
337
+ | 234 | blk.25.ffn_down.weight | 0x16e1f7720 | 0x3b80000 |
338
+ | 235 | blk.25.ffn_gate.weight | 0x171d77720 | 0x3b80000 |
339
+ | 236 | blk.25.ffn_norm.weight | 0x1758f7720 | 0x4000 |
340
+ | 237 | blk.25.ffn_up.weight | 0x1758fb720 | 0x3b80000 |
341
+ | 238 | blk.26.attn_k.weight | 0x17947b720 | 0x440000 |
342
+ | 239 | blk.26.attn_norm.weight | 0x1798bb720 | 0x4000 |
343
+ | 240 | blk.26.attn_output.weight | 0x1798bf720 | 0x1100000 |
344
+ | 241 | blk.26.attn_q.weight | 0x17a9bf720 | 0x1100000 |
345
+ | 242 | blk.26.attn_v.weight | 0x17babf720 | 0x800000 |
346
+ | 243 | blk.26.ffn_down.weight | 0x17c2bf720 | 0x3b80000 |
347
+ | 244 | blk.26.ffn_gate.weight | 0x17fe3f720 | 0x3b80000 |
348
+ | 245 | blk.26.ffn_norm.weight | 0x1839bf720 | 0x4000 |
349
+ | 246 | blk.26.ffn_up.weight | 0x1839c3720 | 0x3b80000 |
350
+ | 247 | blk.27.attn_k.weight | 0x187543720 | 0x440000 |
351
+ | 248 | blk.27.attn_norm.weight | 0x187983720 | 0x4000 |
352
+ | 249 | blk.27.attn_output.weight | 0x187987720 | 0x1100000 |
353
+ | 250 | blk.27.attn_q.weight | 0x188a87720 | 0x1100000 |
354
+ | 251 | blk.27.attn_v.weight | 0x189b87720 | 0x800000 |
355
+ | 252 | blk.27.ffn_down.weight | 0x18a387720 | 0x3b80000 |
356
+ | 253 | blk.27.ffn_gate.weight | 0x18df07720 | 0x3b80000 |
357
+ | 254 | blk.27.ffn_norm.weight | 0x191a87720 | 0x4000 |
358
+ | 255 | blk.27.ffn_up.weight | 0x191a8b720 | 0x3b80000 |
359
+ | 256 | blk.28.attn_k.weight | 0x19560b720 | 0x440000 |
360
+ | 257 | blk.28.attn_norm.weight | 0x195a4b720 | 0x4000 |
361
+ | 258 | blk.28.attn_output.weight | 0x195a4f720 | 0x1100000 |
362
+ | 259 | blk.28.attn_q.weight | 0x196b4f720 | 0x1100000 |
363
+ | 260 | blk.28.attn_v.weight | 0x197c4f720 | 0x800000 |
364
+ | 261 | blk.28.ffn_down.weight | 0x19844f720 | 0x3b80000 |
365
+ | 262 | blk.28.ffn_gate.weight | 0x19bfcf720 | 0x3b80000 |
366
+ | 263 | blk.28.ffn_norm.weight | 0x19fb4f720 | 0x4000 |
367
+ | 264 | blk.28.ffn_up.weight | 0x19fb53720 | 0x3b80000 |
368
+ | 265 | blk.29.attn_k.weight | 0x1a36d3720 | 0x440000 |
369
+ | 266 | blk.29.attn_norm.weight | 0x1a3b13720 | 0x4000 |
370
+ | 267 | blk.29.attn_output.weight | 0x1a3b17720 | 0x1100000 |
371
+ | 268 | blk.29.attn_q.weight | 0x1a4c17720 | 0x1100000 |
372
+ | 269 | blk.29.attn_v.weight | 0x1a5d17720 | 0x800000 |
373
+ | 270 | blk.29.ffn_down.weight | 0x1a6517720 | 0x3b80000 |
374
+ | 271 | blk.29.ffn_gate.weight | 0x1aa097720 | 0x3b80000 |
375
+ | 272 | blk.29.ffn_norm.weight | 0x1adc17720 | 0x4000 |
376
+ | 273 | blk.29.ffn_up.weight | 0x1adc1b720 | 0x3b80000 |
377
+ | 274 | blk.30.attn_k.weight | 0x1b179b720 | 0x440000 |
378
+ | 275 | blk.30.attn_norm.weight | 0x1b1bdb720 | 0x4000 |
379
+ | 276 | blk.30.attn_output.weight | 0x1b1bdf720 | 0x1100000 |
380
+ | 277 | blk.30.attn_q.weight | 0x1b2cdf720 | 0x1100000 |
381
+ | 278 | blk.30.attn_v.weight | 0x1b3ddf720 | 0x800000 |
382
+ | 279 | blk.30.ffn_down.weight | 0x1b45df720 | 0x3b80000 |
383
+ | 280 | blk.30.ffn_gate.weight | 0x1b815f720 | 0x3b80000 |
384
+ | 281 | blk.30.ffn_norm.weight | 0x1bbcdf720 | 0x4000 |
385
+ | 282 | blk.30.ffn_up.weight | 0x1bbce3720 | 0x3b80000 |
386
+ | 283 | blk.31.attn_k.weight | 0x1bf863720 | 0x348000 |
387
+ | 284 | blk.31.attn_norm.weight | 0x1bfbab720 | 0x4000 |
388
+ | 285 | blk.31.attn_output.weight | 0x1bfbaf720 | 0x1100000 |
389
+ | 286 | blk.31.attn_q.weight | 0x1c0caf720 | 0xd20000 |
390
+ | 287 | blk.31.attn_v.weight | 0x1c19cf720 | 0x440000 |
391
+ | 288 | blk.31.ffn_down.weight | 0x1c1e0f720 | 0x3b80000 |
392
+ | 289 | blk.31.ffn_gate.weight | 0x1c598f720 | 0x3b80000 |
393
+ | 290 | blk.31.ffn_norm.weight | 0x1c950f720 | 0x4000 |
394
+ | 291 | blk.31.ffn_up.weight | 0x1c9513720 | 0x3b80000 |
395
+
396
+ ### <a name="base">Base Tensor Group : ~1B Elements</a>
397
+
398
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
399
+ |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----|
400
+ | 0 | output.weight | Output (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q8_0 |
401
+ | 1 | output_norm.weight | Output Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
402
+ | 2 | rope_freqs.weight | Rope_Freqs (W) | ( 64) 64 | 64 x 1 x 1 x 1 | F32 |
403
+ | 3 | token_embd.weight | Token Embedding (W) | (~525M) 525336576 | 4096 x 128256 x 1 x 1 | Q3_K |
404
+
405
+ - Total elements in base: ( ~1B) 1050677312
406
+ - Percentage of total elements: 13.08%
407
+
408
+
409
+ ### <a name="blk_0">Block 0 Tensor Group : ~218M Elements</a>
410
+
411
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
412
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
413
+ | 4 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
414
+ | 5 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
415
+ | 6 | blk.0.attn_output.weight | Block 0 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
416
+ | 7 | blk.0.attn_q.weight | Block 0 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
417
+ | 8 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
418
+ | 9 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
419
+ | 10 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
420
+ | 11 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
421
+ | 12 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
422
+
423
+ - Total elements in blk.0: (~218M) 218112000
424
+ - Percentage of total elements: 2.72%
425
+
426
+
427
+ ### <a name="blk_1">Block 1 Tensor Group : ~218M Elements</a>
428
+
429
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
430
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
431
+ | 13 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
432
+ | 14 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
433
+ | 15 | blk.1.attn_output.weight | Block 1 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
434
+ | 16 | blk.1.attn_q.weight | Block 1 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
435
+ | 17 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
436
+ | 18 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
437
+ | 19 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
438
+ | 20 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
439
+ | 21 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
440
+
441
+ - Total elements in blk.1: (~218M) 218112000
442
+ - Percentage of total elements: 2.72%
443
+
444
+
445
+ ### <a name="blk_2">Block 2 Tensor Group : ~218M Elements</a>
446
+
447
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
448
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
449
+ | 22 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
450
+ | 23 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
451
+ | 24 | blk.2.attn_output.weight | Block 2 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
452
+ | 25 | blk.2.attn_q.weight | Block 2 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
453
+ | 26 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
454
+ | 27 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
455
+ | 28 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
456
+ | 29 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
457
+ | 30 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
458
+
459
+ - Total elements in blk.2: (~218M) 218112000
460
+ - Percentage of total elements: 2.72%
461
+
462
+
463
+ ### <a name="blk_3">Block 3 Tensor Group : ~218M Elements</a>
464
+
465
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
466
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
467
+ | 31 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
468
+ | 32 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
469
+ | 33 | blk.3.attn_output.weight | Block 3 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
470
+ | 34 | blk.3.attn_q.weight | Block 3 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
471
+ | 35 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
472
+ | 36 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
473
+ | 37 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
474
+ | 38 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
475
+ | 39 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
476
+
477
+ - Total elements in blk.3: (~218M) 218112000
478
+ - Percentage of total elements: 2.72%
479
+
480
+
481
+ ### <a name="blk_4">Block 4 Tensor Group : ~218M Elements</a>
482
+
483
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
484
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
485
+ | 40 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
486
+ | 41 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
487
+ | 42 | blk.4.attn_output.weight | Block 4 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
488
+ | 43 | blk.4.attn_q.weight | Block 4 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
489
+ | 44 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
490
+ | 45 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
491
+ | 46 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
492
+ | 47 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
493
+ | 48 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
494
+
495
+ - Total elements in blk.4: (~218M) 218112000
496
+ - Percentage of total elements: 2.72%
497
+
498
+
499
+ ### <a name="blk_5">Block 5 Tensor Group : ~218M Elements</a>
500
+
501
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
502
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
503
+ | 49 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
504
+ | 50 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
505
+ | 51 | blk.5.attn_output.weight | Block 5 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
506
+ | 52 | blk.5.attn_q.weight | Block 5 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
507
+ | 53 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
508
+ | 54 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
509
+ | 55 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
510
+ | 56 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
511
+ | 57 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
512
+
513
+ - Total elements in blk.5: (~218M) 218112000
514
+ - Percentage of total elements: 2.72%
515
+
516
+
517
+ ### <a name="blk_6">Block 6 Tensor Group : ~218M Elements</a>
518
+
519
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
520
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
521
+ | 58 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
522
+ | 59 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
523
+ | 60 | blk.6.attn_output.weight | Block 6 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
524
+ | 61 | blk.6.attn_q.weight | Block 6 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
525
+ | 62 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
526
+ | 63 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
527
+ | 64 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
528
+ | 65 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
529
+ | 66 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
530
+
531
+ - Total elements in blk.6: (~218M) 218112000
532
+ - Percentage of total elements: 2.72%
533
+
534
+
535
+ ### <a name="blk_7">Block 7 Tensor Group : ~218M Elements</a>
536
+
537
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
538
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
539
+ | 67 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
540
+ | 68 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
541
+ | 69 | blk.7.attn_output.weight | Block 7 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
542
+ | 70 | blk.7.attn_q.weight | Block 7 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
543
+ | 71 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
544
+ | 72 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
545
+ | 73 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
546
+ | 74 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
547
+ | 75 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
548
+
549
+ - Total elements in blk.7: (~218M) 218112000
550
+ - Percentage of total elements: 2.72%
551
+
552
+
553
+ ### <a name="blk_8">Block 8 Tensor Group : ~218M Elements</a>
554
+
555
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
556
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
557
+ | 76 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
558
+ | 77 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
559
+ | 78 | blk.8.attn_output.weight | Block 8 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
560
+ | 79 | blk.8.attn_q.weight | Block 8 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
561
+ | 80 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
562
+ | 81 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
563
+ | 82 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
564
+ | 83 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
565
+ | 84 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
566
+
567
+ - Total elements in blk.8: (~218M) 218112000
568
+ - Percentage of total elements: 2.72%
569
+
570
+
571
+ ### <a name="blk_9">Block 9 Tensor Group : ~218M Elements</a>
572
+
573
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
574
+ |-----:|:-------------------------|:-----------------------------------------------|:----------------|:----------------------|:-----|
575
+ | 85 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
576
+ | 86 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
577
+ | 87 | blk.9.attn_output.weight | Block 9 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
578
+ | 88 | blk.9.attn_q.weight | Block 9 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
579
+ | 89 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
580
+ | 90 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
581
+ | 91 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
582
+ | 92 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
583
+ | 93 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
584
+
585
+ - Total elements in blk.9: (~218M) 218112000
586
+ - Percentage of total elements: 2.72%
587
+
588
+
589
+ ### <a name="blk_10">Block 10 Tensor Group : ~218M Elements</a>
590
+
591
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
592
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
593
+ | 94 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
594
+ | 95 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
595
+ | 96 | blk.10.attn_output.weight | Block 10 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
596
+ | 97 | blk.10.attn_q.weight | Block 10 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
597
+ | 98 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
598
+ | 99 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
599
+ | 100 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
600
+ | 101 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
601
+ | 102 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
602
+
603
+ - Total elements in blk.10: (~218M) 218112000
604
+ - Percentage of total elements: 2.72%
605
+
606
+
607
+ ### <a name="blk_11">Block 11 Tensor Group : ~218M Elements</a>
608
+
609
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
610
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
611
+ | 103 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
612
+ | 104 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
613
+ | 105 | blk.11.attn_output.weight | Block 11 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
614
+ | 106 | blk.11.attn_q.weight | Block 11 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
615
+ | 107 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
616
+ | 108 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
617
+ | 109 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
618
+ | 110 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
619
+ | 111 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
620
+
621
+ - Total elements in blk.11: (~218M) 218112000
622
+ - Percentage of total elements: 2.72%
623
+
624
+
625
+ ### <a name="blk_12">Block 12 Tensor Group : ~218M Elements</a>
626
+
627
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
628
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
629
+ | 112 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
630
+ | 113 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
631
+ | 114 | blk.12.attn_output.weight | Block 12 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
632
+ | 115 | blk.12.attn_q.weight | Block 12 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
633
+ | 116 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
634
+ | 117 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
635
+ | 118 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
636
+ | 119 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
637
+ | 120 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
638
+
639
+ - Total elements in blk.12: (~218M) 218112000
640
+ - Percentage of total elements: 2.72%
641
+
642
+
643
+ ### <a name="blk_13">Block 13 Tensor Group : ~218M Elements</a>
644
+
645
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
646
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
647
+ | 121 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
648
+ | 122 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
649
+ | 123 | blk.13.attn_output.weight | Block 13 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
650
+ | 124 | blk.13.attn_q.weight | Block 13 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
651
+ | 125 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
652
+ | 126 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
653
+ | 127 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
654
+ | 128 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
655
+ | 129 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
656
+
657
+ - Total elements in blk.13: (~218M) 218112000
658
+ - Percentage of total elements: 2.72%
659
+
660
+
661
+ ### <a name="blk_14">Block 14 Tensor Group : ~218M Elements</a>
662
+
663
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
664
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
665
+ | 130 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
666
+ | 131 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
667
+ | 132 | blk.14.attn_output.weight | Block 14 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
668
+ | 133 | blk.14.attn_q.weight | Block 14 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
669
+ | 134 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
670
+ | 135 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
671
+ | 136 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
672
+ | 137 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
673
+ | 138 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
674
+
675
+ - Total elements in blk.14: (~218M) 218112000
676
+ - Percentage of total elements: 2.72%
677
+
678
+
679
+ ### <a name="blk_15">Block 15 Tensor Group : ~218M Elements</a>
680
+
681
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
682
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
683
+ | 139 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
684
+ | 140 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
685
+ | 141 | blk.15.attn_output.weight | Block 15 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
686
+ | 142 | blk.15.attn_q.weight | Block 15 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
687
+ | 143 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
688
+ | 144 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
689
+ | 145 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
690
+ | 146 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
691
+ | 147 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q6_K |
692
+
693
+ - Total elements in blk.15: (~218M) 218112000
694
+ - Percentage of total elements: 2.72%
695
+
696
+
697
+ ### <a name="blk_16">Block 16 Tensor Group : ~218M Elements</a>
698
+
699
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
700
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
701
+ | 148 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
702
+ | 149 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
703
+ | 150 | blk.16.attn_output.weight | Block 16 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
704
+ | 151 | blk.16.attn_q.weight | Block 16 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
705
+ | 152 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
706
+ | 153 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
707
+ | 154 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
708
+ | 155 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
709
+ | 156 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
710
+
711
+ - Total elements in blk.16: (~218M) 218112000
712
+ - Percentage of total elements: 2.72%
713
+
714
+
715
+ ### <a name="blk_17">Block 17 Tensor Group : ~218M Elements</a>
716
+
717
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
718
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
719
+ | 157 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
720
+ | 158 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
721
+ | 159 | blk.17.attn_output.weight | Block 17 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
722
+ | 160 | blk.17.attn_q.weight | Block 17 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
723
+ | 161 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
724
+ | 162 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
725
+ | 163 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
726
+ | 164 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
727
+ | 165 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
728
+
729
+ - Total elements in blk.17: (~218M) 218112000
730
+ - Percentage of total elements: 2.72%
731
+
732
+
733
+ ### <a name="blk_18">Block 18 Tensor Group : ~218M Elements</a>
734
+
735
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
736
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
737
+ | 166 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
738
+ | 167 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
739
+ | 168 | blk.18.attn_output.weight | Block 18 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
740
+ | 169 | blk.18.attn_q.weight | Block 18 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
741
+ | 170 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
742
+ | 171 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
743
+ | 172 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
744
+ | 173 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
745
+ | 174 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
746
+
747
+ - Total elements in blk.18: (~218M) 218112000
748
+ - Percentage of total elements: 2.72%
749
+
750
+
751
+ ### <a name="blk_19">Block 19 Tensor Group : ~218M Elements</a>
752
+
753
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
754
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
755
+ | 175 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
756
+ | 176 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
757
+ | 177 | blk.19.attn_output.weight | Block 19 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
758
+ | 178 | blk.19.attn_q.weight | Block 19 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
759
+ | 179 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
760
+ | 180 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
761
+ | 181 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
762
+ | 182 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
763
+ | 183 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
764
+
765
+ - Total elements in blk.19: (~218M) 218112000
766
+ - Percentage of total elements: 2.72%
767
+
768
+
769
+ ### <a name="blk_20">Block 20 Tensor Group : ~218M Elements</a>
770
+
771
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
772
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
773
+ | 184 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
774
+ | 185 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
775
+ | 186 | blk.20.attn_output.weight | Block 20 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
776
+ | 187 | blk.20.attn_q.weight | Block 20 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
777
+ | 188 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
778
+ | 189 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
779
+ | 190 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
780
+ | 191 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
781
+ | 192 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
782
+
783
+ - Total elements in blk.20: (~218M) 218112000
784
+ - Percentage of total elements: 2.72%
785
+
786
+
787
+ ### <a name="blk_21">Block 21 Tensor Group : ~218M Elements</a>
788
+
789
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
790
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
791
+ | 193 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
792
+ | 194 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
793
+ | 195 | blk.21.attn_output.weight | Block 21 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
794
+ | 196 | blk.21.attn_q.weight | Block 21 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
795
+ | 197 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
796
+ | 198 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
797
+ | 199 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
798
+ | 200 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
799
+ | 201 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
800
+
801
+ - Total elements in blk.21: (~218M) 218112000
802
+ - Percentage of total elements: 2.72%
803
+
804
+
805
+ ### <a name="blk_22">Block 22 Tensor Group : ~218M Elements</a>
806
+
807
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
808
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
809
+ | 202 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
810
+ | 203 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
811
+ | 204 | blk.22.attn_output.weight | Block 22 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
812
+ | 205 | blk.22.attn_q.weight | Block 22 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
813
+ | 206 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
814
+ | 207 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
815
+ | 208 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
816
+ | 209 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
817
+ | 210 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
818
+
819
+ - Total elements in blk.22: (~218M) 218112000
820
+ - Percentage of total elements: 2.72%
821
+
822
+
823
+ ### <a name="blk_23">Block 23 Tensor Group : ~218M Elements</a>
824
+
825
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
826
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
827
+ | 211 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
828
+ | 212 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
829
+ | 213 | blk.23.attn_output.weight | Block 23 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
830
+ | 214 | blk.23.attn_q.weight | Block 23 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
831
+ | 215 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
832
+ | 216 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
833
+ | 217 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
834
+ | 218 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
835
+ | 219 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
836
+
837
+ - Total elements in blk.23: (~218M) 218112000
838
+ - Percentage of total elements: 2.72%
839
+
840
+
841
+ ### <a name="blk_24">Block 24 Tensor Group : ~218M Elements</a>
842
+
843
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
844
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
845
+ | 220 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
846
+ | 221 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
847
+ | 222 | blk.24.attn_output.weight | Block 24 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
848
+ | 223 | blk.24.attn_q.weight | Block 24 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
849
+ | 224 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
850
+ | 225 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
851
+ | 226 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
852
+ | 227 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
853
+ | 228 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
854
+
855
+ - Total elements in blk.24: (~218M) 218112000
856
+ - Percentage of total elements: 2.72%
857
+
858
+
859
+ ### <a name="blk_25">Block 25 Tensor Group : ~218M Elements</a>
860
+
861
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
862
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
863
+ | 229 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
864
+ | 230 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
865
+ | 231 | blk.25.attn_output.weight | Block 25 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
866
+ | 232 | blk.25.attn_q.weight | Block 25 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
867
+ | 233 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
868
+ | 234 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
869
+ | 235 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
870
+ | 236 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
871
+ | 237 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
872
+
873
+ - Total elements in blk.25: (~218M) 218112000
874
+ - Percentage of total elements: 2.72%
875
+
876
+
877
+ ### <a name="blk_26">Block 26 Tensor Group : ~218M Elements</a>
878
+
879
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
880
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
881
+ | 238 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
882
+ | 239 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
883
+ | 240 | blk.26.attn_output.weight | Block 26 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
884
+ | 241 | blk.26.attn_q.weight | Block 26 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
885
+ | 242 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
886
+ | 243 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
887
+ | 244 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
888
+ | 245 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
889
+ | 246 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
890
+
891
+ - Total elements in blk.26: (~218M) 218112000
892
+ - Percentage of total elements: 2.72%
893
+
894
+
895
+ ### <a name="blk_27">Block 27 Tensor Group : ~218M Elements</a>
896
+
897
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
898
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
899
+ | 247 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
900
+ | 248 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
901
+ | 249 | blk.27.attn_output.weight | Block 27 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
902
+ | 250 | blk.27.attn_q.weight | Block 27 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
903
+ | 251 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
904
+ | 252 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
905
+ | 253 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
906
+ | 254 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
907
+ | 255 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
908
+
909
+ - Total elements in blk.27: (~218M) 218112000
910
+ - Percentage of total elements: 2.72%
911
+
912
+
913
+ ### <a name="blk_28">Block 28 Tensor Group : ~218M Elements</a>
914
+
915
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
916
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
917
+ | 256 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
918
+ | 257 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
919
+ | 258 | blk.28.attn_output.weight | Block 28 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
920
+ | 259 | blk.28.attn_q.weight | Block 28 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
921
+ | 260 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
922
+ | 261 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
923
+ | 262 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
924
+ | 263 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
925
+ | 264 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
926
+
927
+ - Total elements in blk.28: (~218M) 218112000
928
+ - Percentage of total elements: 2.72%
929
+
930
+
931
+ ### <a name="blk_29">Block 29 Tensor Group : ~218M Elements</a>
932
+
933
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
934
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
935
+ | 265 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
936
+ | 266 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
937
+ | 267 | blk.29.attn_output.weight | Block 29 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
938
+ | 268 | blk.29.attn_q.weight | Block 29 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
939
+ | 269 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
940
+ | 270 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
941
+ | 271 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
942
+ | 272 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
943
+ | 273 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
944
+
945
+ - Total elements in blk.29: (~218M) 218112000
946
+ - Percentage of total elements: 2.72%
947
+
948
+
949
+ ### <a name="blk_30">Block 30 Tensor Group : ~218M Elements</a>
950
+
951
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
952
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
953
+ | 274 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
954
+ | 275 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
955
+ | 276 | blk.30.attn_output.weight | Block 30 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
956
+ | 277 | blk.30.attn_q.weight | Block 30 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
957
+ | 278 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | F16 |
958
+ | 279 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
959
+ | 280 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
960
+ | 281 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
961
+ | 282 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
962
+
963
+ - Total elements in blk.30: (~218M) 218112000
964
+ - Percentage of total elements: 2.72%
965
+
966
+
967
+ ### <a name="blk_31">Block 31 Tensor Group : ~218M Elements</a>
968
+
969
+ | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |
970
+ |-----:|:--------------------------|:------------------------------------------------|:----------------|:----------------------|:-----|
971
+ | 283 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q6_K |
972
+ | 284 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
973
+ | 285 | blk.31.attn_output.weight | Block 31 Attention Output (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q8_0 |
974
+ | 286 | blk.31.attn_q.weight | Block 31 Attention Query (W) | (~17M) 16777216 | 4096 x 4096 x 1 x 1 | Q6_K |
975
+ | 287 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~4M) 4194304 | 4096 x 1024 x 1 x 1 | Q8_0 |
976
+ | 288 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~59M) 58720256 | 14336 x 4096 x 1 x 1 | Q8_0 |
977
+ | 289 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
978
+ | 290 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~4K) 4096 | 4096 x 1 x 1 x 1 | F32 |
979
+ | 291 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~59M) 58720256 | 4096 x 14336 x 1 x 1 | Q8_0 |
980
+
981
+ - Total elements in blk.31: (~218M) 218112000
982
+ - Percentage of total elements: 2.72%