kev-4b: jaredpalmer/kev-4b@4bc64c6b4c4881148661ffb823ce21fcfdc79a0e
Browse files- kev-4b/manifest.json +3 -1
kev-4b/manifest.json
CHANGED
|
@@ -1,6 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"name": "kev-4b",
|
| 3 |
"run": "jaredpalmer/kev-4b@4bc64c6b4c4881148661ffb823ce21fcfdc79a0e",
|
|
|
|
| 4 |
"base": "Qwen/Qwen3.5-4B-Base",
|
| 5 |
"hidden_size": 2560,
|
| 6 |
"head_dim": 256,
|
|
@@ -2029,7 +2030,8 @@
|
|
| 2029 |
"max_abs_dp": 0.062458,
|
| 2030 |
"argmax_flips": 1,
|
| 2031 |
"questions": 29
|
| 2032 |
-
}
|
|
|
|
| 2033 |
}
|
| 2034 |
}
|
| 2035 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"name": "kev-4b",
|
| 3 |
"run": "jaredpalmer/kev-4b@4bc64c6b4c4881148661ffb823ce21fcfdc79a0e",
|
| 4 |
+
"revision": "77bd732ae9a747af",
|
| 5 |
"base": "Qwen/Qwen3.5-4B-Base",
|
| 6 |
"hidden_size": 2560,
|
| 7 |
"head_dim": 256,
|
|
|
|
| 2030 |
"max_abs_dp": 0.062458,
|
| 2031 |
"argmax_flips": 1,
|
| 2032 |
"questions": 29
|
| 2033 |
+
},
|
| 2034 |
+
"max_positions": 8192
|
| 2035 |
}
|
| 2036 |
}
|
| 2037 |
}
|