Upload DistillationWrapper
Browse files- config.json +1 -2
config.json
CHANGED
|
@@ -5,7 +5,7 @@
|
|
| 5 |
],
|
| 6 |
"attention_probs_dropout_prob": 0.1,
|
| 7 |
"classifier_dropout": null,
|
| 8 |
-
"dtype": "
|
| 9 |
"embedding_size": 128,
|
| 10 |
"expert_intermediate_size": 2624,
|
| 11 |
"group_depth": 4,
|
|
@@ -24,7 +24,6 @@
|
|
| 24 |
"num_expert_modules": 3,
|
| 25 |
"num_experts": 8,
|
| 26 |
"num_hidden_layers": 16,
|
| 27 |
-
"output_hidden_states": true,
|
| 28 |
"pad_token_id": 0,
|
| 29 |
"router_jitter_noise": 0.01,
|
| 30 |
"top_k": 2,
|
|
|
|
| 5 |
],
|
| 6 |
"attention_probs_dropout_prob": 0.1,
|
| 7 |
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
"embedding_size": 128,
|
| 10 |
"expert_intermediate_size": 2624,
|
| 11 |
"group_depth": 4,
|
|
|
|
| 24 |
"num_expert_modules": 3,
|
| 25 |
"num_experts": 8,
|
| 26 |
"num_hidden_layers": 16,
|
|
|
|
| 27 |
"pad_token_id": 0,
|
| 28 |
"router_jitter_noise": 0.01,
|
| 29 |
"top_k": 2,
|