Update config.json
Browse files- config.json +6 -6
config.json
CHANGED
|
@@ -4,18 +4,18 @@
|
|
| 4 |
],
|
| 5 |
"attention_bias": false,
|
| 6 |
"attention_dropout": 0.0,
|
| 7 |
-
"bos_token_id":
|
| 8 |
"classifier_activation": "gelu",
|
| 9 |
"classifier_bias": false,
|
| 10 |
"classifier_dropout": 0.0,
|
| 11 |
"classifier_pooling": "mean",
|
| 12 |
-
"cls_token_id":
|
| 13 |
"decoder_bias": true,
|
| 14 |
"deterministic_flash_attn": false,
|
| 15 |
"embedding_dropout": 0.0,
|
| 16 |
-
"eos_token_id":
|
| 17 |
"global_attn_every_n_layers": 3,
|
| 18 |
-
"global_rope_theta":
|
| 19 |
"gradient_checkpointing": false,
|
| 20 |
"hidden_activation": "gelu",
|
| 21 |
"hidden_size": 768,
|
|
@@ -33,10 +33,10 @@
|
|
| 33 |
"norm_eps": 1e-05,
|
| 34 |
"num_attention_heads": 12,
|
| 35 |
"num_hidden_layers": 22,
|
| 36 |
-
"pad_token_id":
|
| 37 |
"position_embedding_type": "absolute",
|
| 38 |
"repad_logits_with_grad": false,
|
| 39 |
-
"sep_token_id":
|
| 40 |
"sparse_pred_ignore_index": -100,
|
| 41 |
"sparse_prediction": false,
|
| 42 |
"torch_dtype": "float32",
|
|
|
|
| 4 |
],
|
| 5 |
"attention_bias": false,
|
| 6 |
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 101,
|
| 8 |
"classifier_activation": "gelu",
|
| 9 |
"classifier_bias": false,
|
| 10 |
"classifier_dropout": 0.0,
|
| 11 |
"classifier_pooling": "mean",
|
| 12 |
+
"cls_token_id": 101,
|
| 13 |
"decoder_bias": true,
|
| 14 |
"deterministic_flash_attn": false,
|
| 15 |
"embedding_dropout": 0.0,
|
| 16 |
+
"eos_token_id": 102,
|
| 17 |
"global_attn_every_n_layers": 3,
|
| 18 |
+
"global_rope_theta": 10000.0,
|
| 19 |
"gradient_checkpointing": false,
|
| 20 |
"hidden_activation": "gelu",
|
| 21 |
"hidden_size": 768,
|
|
|
|
| 33 |
"norm_eps": 1e-05,
|
| 34 |
"num_attention_heads": 12,
|
| 35 |
"num_hidden_layers": 22,
|
| 36 |
+
"pad_token_id": 0,
|
| 37 |
"position_embedding_type": "absolute",
|
| 38 |
"repad_logits_with_grad": false,
|
| 39 |
+
"sep_token_id": 102,
|
| 40 |
"sparse_pred_ignore_index": -100,
|
| 41 |
"sparse_prediction": false,
|
| 42 |
"torch_dtype": "float32",
|