Update Mistral-NeMo-12B-Base-HF/config.json
Browse files
Mistral-NeMo-12B-Base-HF/config.json
CHANGED
@@ -5,11 +5,11 @@
|
|
5 |
"attention_dropout": 0.0,
|
6 |
"bos_token_id": 1,
|
7 |
"eos_token_id": 2,
|
|
|
8 |
"hidden_act": "silu",
|
9 |
"hidden_size": 5120,
|
10 |
"initializer_range": 0.02,
|
11 |
"intermediate_size": 14336,
|
12 |
-
"kv_channels": 128,
|
13 |
"max_position_embeddings": 1024000,
|
14 |
"model_type": "mistral",
|
15 |
"num_attention_heads": 32,
|
@@ -20,7 +20,7 @@
|
|
20 |
"sliding_window": null,
|
21 |
"tie_word_embeddings": false,
|
22 |
"torch_dtype": "bfloat16",
|
23 |
-
"transformers_version": "4.
|
24 |
"use_cache": true,
|
25 |
"vocab_size": 131072
|
26 |
}
|
|
|
5 |
"attention_dropout": 0.0,
|
6 |
"bos_token_id": 1,
|
7 |
"eos_token_id": 2,
|
8 |
+
"head_dim": 128,
|
9 |
"hidden_act": "silu",
|
10 |
"hidden_size": 5120,
|
11 |
"initializer_range": 0.02,
|
12 |
"intermediate_size": 14336,
|
|
|
13 |
"max_position_embeddings": 1024000,
|
14 |
"model_type": "mistral",
|
15 |
"num_attention_heads": 32,
|
|
|
20 |
"sliding_window": null,
|
21 |
"tie_word_embeddings": false,
|
22 |
"torch_dtype": "bfloat16",
|
23 |
+
"transformers_version": "4.43.0.dev0",
|
24 |
"use_cache": true,
|
25 |
"vocab_size": 131072
|
26 |
}
|