Update config.json
Browse files- config.json +9 -7
config.json
CHANGED
@@ -1,18 +1,20 @@
|
|
1 |
{
|
2 |
"_name_or_path": "adowu/astral-128k",
|
3 |
-
"architectures": [
|
|
|
|
|
4 |
"attention_dropout": 0.1,
|
5 |
"bos_token_id": 1,
|
6 |
"eos_token_id": 2,
|
7 |
"hidden_act": "silu",
|
8 |
-
"hidden_size":
|
9 |
"initializer_range": 0.02,
|
10 |
-
"intermediate_size": 8192,
|
11 |
"max_position_embeddings": 128000,
|
12 |
"model_type": "mistral",
|
13 |
-
"num_attention_heads": 32,
|
14 |
-
"num_hidden_layers": 16,
|
15 |
-
"num_key_value_heads": 8,
|
16 |
"pad_token_id": 2,
|
17 |
"rms_norm_eps": 1e-05,
|
18 |
"rope_theta": 10000.0,
|
@@ -21,5 +23,5 @@
|
|
21 |
"torch_dtype": "bfloat16",
|
22 |
"transformers_version": "4.39.3",
|
23 |
"use_cache": true,
|
24 |
-
"vocab_size": 32002
|
25 |
}
|
|
|
1 |
{
|
2 |
"_name_or_path": "adowu/astral-128k",
|
3 |
+
"architectures": [
|
4 |
+
"MistralForCausalLM"
|
5 |
+
],
|
6 |
"attention_dropout": 0.1,
|
7 |
"bos_token_id": 1,
|
8 |
"eos_token_id": 2,
|
9 |
"hidden_act": "silu",
|
10 |
+
"hidden_size": 1024,
|
11 |
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 8192,
|
13 |
"max_position_embeddings": 128000,
|
14 |
"model_type": "mistral",
|
15 |
+
"num_attention_heads": 32,
|
16 |
+
"num_hidden_layers": 16,
|
17 |
+
"num_key_value_heads": 8,
|
18 |
"pad_token_id": 2,
|
19 |
"rms_norm_eps": 1e-05,
|
20 |
"rope_theta": 10000.0,
|
|
|
23 |
"torch_dtype": "bfloat16",
|
24 |
"transformers_version": "4.39.3",
|
25 |
"use_cache": true,
|
26 |
+
"vocab_size": 32002
|
27 |
}
|