Initial GPTQ model commit
Browse files- config.json +40 -0
config.json
ADDED
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"architectures": [
|
4 |
+
"BaichuanForCausalLM"
|
5 |
+
],
|
6 |
+
"auto_map": {
|
7 |
+
"AutoConfig": "configuration_baichuan.BaichuanConfig",
|
8 |
+
"AutoModelForCausalLM": "modeling_baichuan.BaichuanForCausalLM"
|
9 |
+
},
|
10 |
+
"bos_token_id": 1,
|
11 |
+
"eos_token_id": 2,
|
12 |
+
"gradient_checkpointing": false,
|
13 |
+
"hidden_act": "silu",
|
14 |
+
"hidden_size": 5120,
|
15 |
+
"initializer_range": 0.02,
|
16 |
+
"intermediate_size": 13696,
|
17 |
+
"model_max_length": 4096,
|
18 |
+
"model_type": "baichuan",
|
19 |
+
"num_attention_heads": 40,
|
20 |
+
"num_hidden_layers": 40,
|
21 |
+
"pad_token_id": 0,
|
22 |
+
"rms_norm_eps": 1e-06,
|
23 |
+
"tie_word_embeddings": false,
|
24 |
+
"torch_dtype": "bfloat16",
|
25 |
+
"transformers_version": "4.29.2",
|
26 |
+
"use_cache": true,
|
27 |
+
"vocab_size": 125696,
|
28 |
+
"pretraining_tp": 1,
|
29 |
+
"quantization_config": {
|
30 |
+
"bits": 4,
|
31 |
+
"group_size": 128,
|
32 |
+
"damp_percent": 0.1,
|
33 |
+
"desc_act": false,
|
34 |
+
"sym": true,
|
35 |
+
"true_sequential": true,
|
36 |
+
"model_name_or_path": null,
|
37 |
+
"model_file_base_name": "model",
|
38 |
+
"quant_method": "gptq"
|
39 |
+
}
|
40 |
+
}
|