kev-0.8b: jaredpalmer/kev-0.8b@225679690cdd1de6fceb1258b1bddf61c493cee9
Browse files- kev-0.8b/manifest.json +7 -3
kev-0.8b/manifest.json
CHANGED
|
@@ -1,6 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"name": "kev-0.8b",
|
| 3 |
"run": "jaredpalmer/kev-0.8b@225679690cdd1de6fceb1258b1bddf61c493cee9",
|
|
|
|
| 4 |
"base": "Qwen/Qwen3.5-0.8B-Base",
|
| 5 |
"hidden_size": 1024,
|
| 6 |
"head_dim": 256,
|
|
@@ -1323,7 +1324,8 @@
|
|
| 1323 |
"max_abs_dp": 0.0967,
|
| 1324 |
"argmax_flips": 0,
|
| 1325 |
"questions": 49
|
| 1326 |
-
}
|
|
|
|
| 1327 |
},
|
| 1328 |
"q8": {
|
| 1329 |
"model": "r-2256796/q8/model.onnx",
|
|
@@ -2629,7 +2631,8 @@
|
|
| 2629 |
"max_abs_dp": 0.068566,
|
| 2630 |
"argmax_flips": 0,
|
| 2631 |
"questions": 49
|
| 2632 |
-
}
|
|
|
|
| 2633 |
},
|
| 2634 |
"fp32": {
|
| 2635 |
"model": "r-2256796/fp32/model.onnx",
|
|
@@ -3881,7 +3884,8 @@
|
|
| 3881 |
"max_abs_dp": 5.8e-05,
|
| 3882 |
"argmax_flips": 0,
|
| 3883 |
"questions": 49
|
| 3884 |
-
}
|
|
|
|
| 3885 |
}
|
| 3886 |
}
|
| 3887 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"name": "kev-0.8b",
|
| 3 |
"run": "jaredpalmer/kev-0.8b@225679690cdd1de6fceb1258b1bddf61c493cee9",
|
| 4 |
+
"revision": "12ef77327717f26c",
|
| 5 |
"base": "Qwen/Qwen3.5-0.8B-Base",
|
| 6 |
"hidden_size": 1024,
|
| 7 |
"head_dim": 256,
|
|
|
|
| 1324 |
"max_abs_dp": 0.0967,
|
| 1325 |
"argmax_flips": 0,
|
| 1326 |
"questions": 49
|
| 1327 |
+
},
|
| 1328 |
+
"max_positions": 8192
|
| 1329 |
},
|
| 1330 |
"q8": {
|
| 1331 |
"model": "r-2256796/q8/model.onnx",
|
|
|
|
| 2631 |
"max_abs_dp": 0.068566,
|
| 2632 |
"argmax_flips": 0,
|
| 2633 |
"questions": 49
|
| 2634 |
+
},
|
| 2635 |
+
"max_positions": 8192
|
| 2636 |
},
|
| 2637 |
"fp32": {
|
| 2638 |
"model": "r-2256796/fp32/model.onnx",
|
|
|
|
| 3884 |
"max_abs_dp": 5.8e-05,
|
| 3885 |
"argmax_flips": 0,
|
| 3886 |
"questions": 49
|
| 3887 |
+
},
|
| 3888 |
+
"max_positions": 262144
|
| 3889 |
}
|
| 3890 |
}
|
| 3891 |
}
|