Initial GPTQ model commit
Browse files- config.json +29 -0
config.json
ADDED
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/workspace/WizardLM-Uncensored-falcon-40b",
|
3 |
+
"alibi": false,
|
4 |
+
"apply_residual_connection_post_layernorm": false,
|
5 |
+
"architectures": [
|
6 |
+
"RWForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"auto_map": {
|
10 |
+
"AutoConfig": "configuration_RW.RWConfig",
|
11 |
+
"AutoModelForCausalLM": "modelling_RW.RWForCausalLM"
|
12 |
+
},
|
13 |
+
"bias": false,
|
14 |
+
"bos_token_id": 1,
|
15 |
+
"eos_token_id": 2,
|
16 |
+
"hidden_dropout": 0.0,
|
17 |
+
"hidden_size": 8192,
|
18 |
+
"initializer_range": 0.02,
|
19 |
+
"layer_norm_epsilon": 1e-05,
|
20 |
+
"model_type": "RefinedWeb",
|
21 |
+
"n_head": 128,
|
22 |
+
"n_head_kv": 8,
|
23 |
+
"n_layer": 60,
|
24 |
+
"parallel_attn": true,
|
25 |
+
"torch_dtype": "float32",
|
26 |
+
"transformers_version": "4.30.0.dev0",
|
27 |
+
"use_cache": true,
|
28 |
+
"vocab_size": 65025
|
29 |
+
}
|