Upload config.json with huggingface_hub
Browse files- config.json +11 -4
config.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
"_name_or_path": "mistralai/Mistral-7B-v0.1",
|
3 |
"architectures": [
|
4 |
-
"
|
5 |
],
|
6 |
"attention_dropout": 0.0,
|
7 |
"bos_token_id": 1,
|
@@ -12,7 +12,7 @@
|
|
12 |
"initializer_range": 0.02,
|
13 |
"intermediate_size": 14336,
|
14 |
"max_position_embeddings": 32768,
|
15 |
-
"model_type": "
|
16 |
"num_attention_heads": 32,
|
17 |
"num_hidden_layers": 32,
|
18 |
"num_key_value_heads": 8,
|
@@ -23,5 +23,12 @@
|
|
23 |
"torch_dtype": "bfloat16",
|
24 |
"transformers_version": "4.44.2",
|
25 |
"use_cache": true,
|
26 |
-
"vocab_size": 32000
|
27 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
{
|
2 |
"_name_or_path": "mistralai/Mistral-7B-v0.1",
|
3 |
"architectures": [
|
4 |
+
"MergedMistralForCausalLM"
|
5 |
],
|
6 |
"attention_dropout": 0.0,
|
7 |
"bos_token_id": 1,
|
|
|
12 |
"initializer_range": 0.02,
|
13 |
"intermediate_size": 14336,
|
14 |
"max_position_embeddings": 32768,
|
15 |
+
"model_type": "mergedmistral",
|
16 |
"num_attention_heads": 32,
|
17 |
"num_hidden_layers": 32,
|
18 |
"num_key_value_heads": 8,
|
|
|
23 |
"torch_dtype": "bfloat16",
|
24 |
"transformers_version": "4.44.2",
|
25 |
"use_cache": true,
|
26 |
+
"vocab_size": 32000,
|
27 |
+
"num_merged_models": 4,
|
28 |
+
"non_linearity": null,
|
29 |
+
"dam_embedding_layer": true,
|
30 |
+
"dam_layernorms": true,
|
31 |
+
"uses_base_model": true,
|
32 |
+
"is_embedding_coef_trainable": false,
|
33 |
+
"is_norm_coef_trainable": false
|
34 |
+
}
|