Update config.json
Browse files- config.json +7 -11
config.json
CHANGED
|
@@ -1,16 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
"architectures": [
|
| 4 |
"LlamaForCausalLM"
|
| 5 |
],
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
| 8 |
-
"bos_token_id": 128000,
|
| 9 |
-
"eos_token_id": [
|
| 10 |
-
128001,
|
| 11 |
-
128008,
|
| 12 |
-
128009
|
| 13 |
-
],
|
| 14 |
"head_dim": 128,
|
| 15 |
"hidden_act": "silu",
|
| 16 |
"hidden_size": 3072,
|
|
@@ -18,7 +17,6 @@
|
|
| 18 |
"intermediate_size": 8192,
|
| 19 |
"max_position_embeddings": 131072,
|
| 20 |
"mlp_bias": false,
|
| 21 |
-
"model_type": "llama",
|
| 22 |
"num_attention_heads": 24,
|
| 23 |
"num_hidden_layers": 28,
|
| 24 |
"num_key_value_heads": 8,
|
|
@@ -33,8 +31,6 @@
|
|
| 33 |
},
|
| 34 |
"rope_theta": 500000.0,
|
| 35 |
"tie_word_embeddings": true,
|
| 36 |
-
"torch_dtype": "float16",
|
| 37 |
"transformers_version": "4.46.3",
|
| 38 |
-
"use_cache": true
|
| 39 |
-
"vocab_size": 128256
|
| 40 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"model_type": "llama",
|
| 3 |
+
"torch_dtype": "float16",
|
| 4 |
+
"eos_token_id": 128009,
|
| 5 |
+
"bos_token_id": 128000,
|
| 6 |
+
"pad_token_id": 128009,
|
| 7 |
+
"vocab_size": 128256,
|
| 8 |
"architectures": [
|
| 9 |
"LlamaForCausalLM"
|
| 10 |
],
|
| 11 |
"attention_bias": false,
|
| 12 |
"attention_dropout": 0.0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
"head_dim": 128,
|
| 14 |
"hidden_act": "silu",
|
| 15 |
"hidden_size": 3072,
|
|
|
|
| 17 |
"intermediate_size": 8192,
|
| 18 |
"max_position_embeddings": 131072,
|
| 19 |
"mlp_bias": false,
|
|
|
|
| 20 |
"num_attention_heads": 24,
|
| 21 |
"num_hidden_layers": 28,
|
| 22 |
"num_key_value_heads": 8,
|
|
|
|
| 31 |
},
|
| 32 |
"rope_theta": 500000.0,
|
| 33 |
"tie_word_embeddings": true,
|
|
|
|
| 34 |
"transformers_version": "4.46.3",
|
| 35 |
+
"use_cache": true
|
|
|
|
| 36 |
}
|