POWERHACK commited on
Commit
4fa830e
1 Parent(s): f678102

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +5 -21
config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_name_or_path": "Qwen/Qwen2.5-3B",
3
  "architectures": [
4
  "Qwen2ForCausalLM"
5
  ],
@@ -16,29 +15,14 @@
16
  "num_attention_heads": 16,
17
  "num_hidden_layers": 36,
18
  "num_key_value_heads": 2,
19
- "quantization_config": {
20
- "_load_in_4bit": false,
21
- "_load_in_8bit": true,
22
- "bnb_4bit_compute_dtype": "float32",
23
- "bnb_4bit_quant_storage": "uint8",
24
- "bnb_4bit_quant_type": "fp4",
25
- "bnb_4bit_use_double_quant": false,
26
- "llm_int8_enable_fp32_cpu_offload": false,
27
- "llm_int8_has_fp16_weight": false,
28
- "llm_int8_skip_modules": null,
29
- "llm_int8_threshold": 6.0,
30
- "load_in_4bit": false,
31
- "load_in_8bit": true,
32
- "quant_method": "bitsandbytes"
33
- },
34
  "rms_norm_eps": 1e-06,
35
  "rope_theta": 1000000.0,
36
- "sliding_window": null,
37
  "tie_word_embeddings": true,
38
  "torch_dtype": "bfloat16",
39
- "transformers_version": "4.44.2",
40
- "use_cache": false,
41
  "use_mrope": false,
42
  "use_sliding_window": false,
43
- "vocab_size": 151665
44
- }
 
1
  {
 
2
  "architectures": [
3
  "Qwen2ForCausalLM"
4
  ],
 
15
  "num_attention_heads": 16,
16
  "num_hidden_layers": 36,
17
  "num_key_value_heads": 2,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
18
  "rms_norm_eps": 1e-06,
19
  "rope_theta": 1000000.0,
20
+ "sliding_window": 32768,
21
  "tie_word_embeddings": true,
22
  "torch_dtype": "bfloat16",
23
+ "transformers_version": "4.40.1",
24
+ "use_cache": true,
25
  "use_mrope": false,
26
  "use_sliding_window": false,
27
+ "vocab_size": 151936
28
+ }