Commit
•
e93b959
1
Parent(s):
8e2162d
uploaded model
Browse files- config.json +44 -0
config.json
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "fairseq-dense-1.3B",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"XGLMForCausalLM"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.1,
|
9 |
+
"attention_heads": 32,
|
10 |
+
"bos_token_id": 50257,
|
11 |
+
"d_model": 2048,
|
12 |
+
"decoder_start_token_id": 2,
|
13 |
+
"dropout": 0.1,
|
14 |
+
"eos_token_id": 50259,
|
15 |
+
"ffn_dim": 8192,
|
16 |
+
"init_std": 0.02,
|
17 |
+
"layerdrop": 0.0,
|
18 |
+
"max_position_embeddings": 2048,
|
19 |
+
"model_type": "xglm",
|
20 |
+
"newlinemode": "s",
|
21 |
+
"num_layers": 24,
|
22 |
+
"pad_token_id": 1,
|
23 |
+
"quantization_config": {
|
24 |
+
"_load_in_4bit": true,
|
25 |
+
"_load_in_8bit": false,
|
26 |
+
"bnb_4bit_compute_dtype": "float32",
|
27 |
+
"bnb_4bit_quant_storage": "uint8",
|
28 |
+
"bnb_4bit_quant_type": "fp4",
|
29 |
+
"bnb_4bit_use_double_quant": false,
|
30 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
31 |
+
"llm_int8_has_fp16_weight": false,
|
32 |
+
"llm_int8_skip_modules": null,
|
33 |
+
"llm_int8_threshold": 6.0,
|
34 |
+
"load_in_4bit": true,
|
35 |
+
"load_in_8bit": false,
|
36 |
+
"quant_method": "bitsandbytes"
|
37 |
+
},
|
38 |
+
"scale_embedding": true,
|
39 |
+
"tokenizer_class": "GPT2Tokenizer",
|
40 |
+
"torch_dtype": "float16",
|
41 |
+
"transformers_version": "4.39.3",
|
42 |
+
"use_cache": true,
|
43 |
+
"vocab_size": 50261
|
44 |
+
}
|