danielhanchen commited on
Commit
947abc4
1 Parent(s): 60c85a8

Upload config

Browse files
Files changed (1) hide show
  1. config.json +2 -17
config.json CHANGED
@@ -6,6 +6,7 @@
6
  "attention_dropout": 0.0,
7
  "bos_token_id": 1,
8
  "eos_token_id": 2,
 
9
  "hidden_act": "silu",
10
  "hidden_size": 4096,
11
  "initializer_range": 0.02,
@@ -15,28 +16,12 @@
15
  "num_attention_heads": 32,
16
  "num_hidden_layers": 32,
17
  "num_key_value_heads": 8,
18
- "pad_token_id": 770,
19
- "quantization_config": {
20
- "_load_in_4bit": true,
21
- "_load_in_8bit": false,
22
- "bnb_4bit_compute_dtype": "bfloat16",
23
- "bnb_4bit_quant_storage": "uint8",
24
- "bnb_4bit_quant_type": "nf4",
25
- "bnb_4bit_use_double_quant": true,
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": true,
31
- "load_in_8bit": false,
32
- "quant_method": "bitsandbytes"
33
- },
34
  "rms_norm_eps": 1e-05,
35
  "rope_theta": 1000000.0,
36
  "sliding_window": null,
37
  "tie_word_embeddings": false,
38
  "torch_dtype": "bfloat16",
39
- "transformers_version": "4.41.0",
40
  "use_cache": true,
41
  "vocab_size": 32768
42
  }
 
6
  "attention_dropout": 0.0,
7
  "bos_token_id": 1,
8
  "eos_token_id": 2,
9
+ "head_dim": 128,
10
  "hidden_act": "silu",
11
  "hidden_size": 4096,
12
  "initializer_range": 0.02,
 
16
  "num_attention_heads": 32,
17
  "num_hidden_layers": 32,
18
  "num_key_value_heads": 8,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
19
  "rms_norm_eps": 1e-05,
20
  "rope_theta": 1000000.0,
21
  "sliding_window": null,
22
  "tie_word_embeddings": false,
23
  "torch_dtype": "bfloat16",
24
+ "transformers_version": "4.44.0",
25
  "use_cache": true,
26
  "vocab_size": 32768
27
  }