Upload DistillationWrapper
Browse files- config.json +1 -1
- model.safetensors +2 -2
config.json
CHANGED
|
@@ -23,7 +23,7 @@
|
|
| 23 |
"num_attention_heads": 16,
|
| 24 |
"num_expert_modules": 2,
|
| 25 |
"num_experts": 8,
|
| 26 |
-
"num_hidden_layers":
|
| 27 |
"pad_token_id": 0,
|
| 28 |
"router_jitter_noise": 0.01,
|
| 29 |
"top_k": 2,
|
|
|
|
| 23 |
"num_attention_heads": 16,
|
| 24 |
"num_expert_modules": 2,
|
| 25 |
"num_experts": 8,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
"pad_token_id": 0,
|
| 28 |
"router_jitter_noise": 0.01,
|
| 29 |
"top_k": 2,
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a54d22fa3969de79efdde14c3ae85f10ba7018b10624f4aabc2e11caa11b18c
|
| 3 |
+
size 255933392
|