mgoin commited on
Commit
82876c9
1 Parent(s): 372b6ff

Updated compression_config to quantization_config

Browse files
Files changed (1) hide show
  1. config.json +25 -25
config.json CHANGED
@@ -8,7 +8,30 @@
8
  "attn_logit_softcapping": 50.0,
9
  "bos_token_id": 2,
10
  "cache_implementation": "hybrid",
11
- "compression_config": {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12
  "config_groups": {
13
  "group_0": {
14
  "input_activations": null,
@@ -43,28 +66,5 @@
43
  "registry_requires_subclass": false,
44
  "sparsity_structure": "unstructured"
45
  }
46
- },
47
- "eos_token_id": 1,
48
- "final_logit_softcapping": 30.0,
49
- "head_dim": 256,
50
- "hidden_act": "gelu_pytorch_tanh",
51
- "hidden_activation": "gelu_pytorch_tanh",
52
- "hidden_size": 3584,
53
- "initializer_range": 0.02,
54
- "intermediate_size": 14336,
55
- "max_position_embeddings": 8192,
56
- "model_type": "gemma2",
57
- "num_attention_heads": 16,
58
- "num_hidden_layers": 42,
59
- "num_key_value_heads": 8,
60
- "pad_token_id": 0,
61
- "query_pre_attn_scalar": 256,
62
- "rms_norm_eps": 1e-06,
63
- "rope_theta": 10000.0,
64
- "sliding_window": 4096,
65
- "sliding_window_size": 4096,
66
- "torch_dtype": "bfloat16",
67
- "transformers_version": "4.43.3",
68
- "use_cache": true,
69
- "vocab_size": 256000
70
  }
 
8
  "attn_logit_softcapping": 50.0,
9
  "bos_token_id": 2,
10
  "cache_implementation": "hybrid",
11
+ "eos_token_id": 1,
12
+ "final_logit_softcapping": 30.0,
13
+ "head_dim": 256,
14
+ "hidden_act": "gelu_pytorch_tanh",
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 3584,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 14336,
19
+ "max_position_embeddings": 8192,
20
+ "model_type": "gemma2",
21
+ "num_attention_heads": 16,
22
+ "num_hidden_layers": 42,
23
+ "num_key_value_heads": 8,
24
+ "pad_token_id": 0,
25
+ "query_pre_attn_scalar": 256,
26
+ "rms_norm_eps": 1e-06,
27
+ "rope_theta": 10000.0,
28
+ "sliding_window": 4096,
29
+ "sliding_window_size": 4096,
30
+ "torch_dtype": "bfloat16",
31
+ "transformers_version": "4.43.3",
32
+ "use_cache": true,
33
+ "vocab_size": 256000,
34
+ "quantization_config": {
35
  "config_groups": {
36
  "group_0": {
37
  "input_activations": null,
 
66
  "registry_requires_subclass": false,
67
  "sparsity_structure": "unstructured"
68
  }
69
+ }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
70
  }