relu activations
Browse files- config.json +2 -3
- generation_config.json +1 -1
- model.safetensors +1 -1
config.json
CHANGED
|
@@ -5,8 +5,8 @@
|
|
| 5 |
"attention_dropout": 0.0,
|
| 6 |
"bos_token_id": 1,
|
| 7 |
"embd_pdrop": 0.0,
|
| 8 |
-
"eos_token_id":
|
| 9 |
-
"hidden_act": "
|
| 10 |
"hidden_size": 64,
|
| 11 |
"initializer_range": 0.02,
|
| 12 |
"intermediate_size": 1,
|
|
@@ -15,7 +15,6 @@
|
|
| 15 |
"num_attention_heads": 1,
|
| 16 |
"num_hidden_layers": 4,
|
| 17 |
"num_key_value_heads": 1,
|
| 18 |
-
"oes_token_id": 2,
|
| 19 |
"original_max_position_embeddings": 4096,
|
| 20 |
"pad_token_id": 3,
|
| 21 |
"resid_pdrop": 0.0,
|
|
|
|
| 5 |
"attention_dropout": 0.0,
|
| 6 |
"bos_token_id": 1,
|
| 7 |
"embd_pdrop": 0.0,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "relu",
|
| 10 |
"hidden_size": 64,
|
| 11 |
"initializer_range": 0.02,
|
| 12 |
"intermediate_size": 1,
|
|
|
|
| 15 |
"num_attention_heads": 1,
|
| 16 |
"num_hidden_layers": 4,
|
| 17 |
"num_key_value_heads": 1,
|
|
|
|
| 18 |
"original_max_position_embeddings": 4096,
|
| 19 |
"pad_token_id": 3,
|
| 20 |
"resid_pdrop": 0.0,
|
generation_config.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
"bos_token_id": 1,
|
| 4 |
-
"eos_token_id":
|
| 5 |
"pad_token_id": 3,
|
| 6 |
"transformers_version": "4.46.1"
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
"pad_token_id": 3,
|
| 6 |
"transformers_version": "4.46.1"
|
| 7 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 142920
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df863f49d09c6cf771c8881f6f8f040b2371e0df4afe018f030cbbe104f0d079
|
| 3 |
size 142920
|