xieli123 commited on
Commit
e0e1a50
·
verified ·
1 Parent(s): 1dc9ae1

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +60 -0
config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Step1ForCausalLM"
4
+ ],
5
+ "auto_map": {
6
+ "AutoConfig": "configuration_step1.Step1Config",
7
+ "AutoModelForCausalLM": "modeling_step1.Step1ForCausalLM"
8
+ },
9
+ "bos_token_id": 1,
10
+ "eos_token_id": 3,
11
+ "hidden_size": 3072,
12
+ "intermediate_size": 8192,
13
+ "max_seq_len": 32768,
14
+ "model_type": "step1",
15
+ "num_attention_groups": 4,
16
+ "num_attention_heads": 48,
17
+ "num_hidden_layers": 32,
18
+ "pad_token_id": 0,
19
+ "quantization_config": {
20
+ "config_groups": {
21
+ "group_0": {
22
+ "format": "pack-quantized",
23
+ "input_activations": null,
24
+ "output_activations": null,
25
+ "targets": [
26
+ "Linear"
27
+ ],
28
+ "weights": {
29
+ "actorder": null,
30
+ "block_structure": null,
31
+ "dynamic": false,
32
+ "group_size": 128,
33
+ "num_bits": 4,
34
+ "observer": "minmax",
35
+ "observer_kwargs": {},
36
+ "strategy": "group",
37
+ "symmetric": true,
38
+ "type": "int"
39
+ }
40
+ }
41
+ },
42
+ "format": "pack-quantized",
43
+ "global_compression_ratio": null,
44
+ "ignore": [
45
+ "lm_head"
46
+ ],
47
+ "kv_cache_scheme": null,
48
+ "quant_method": "compressed-tensors",
49
+ "quantization_status": "compressed",
50
+ "sparsity_config": {},
51
+ "transform_config": {},
52
+ "version": "0.11.0"
53
+ },
54
+ "rms_norm_eps": 1e-05,
55
+ "tie_word_embeddings": false,
56
+ "torch_dtype": "bfloat16",
57
+ "transformers_version": "4.53.3",
58
+ "use_cache": true,
59
+ "vocab_size": 74752
60
+ }