thiera1 commited on
Commit
965f562
·
verified ·
1 Parent(s): 2e38e8d

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +41 -0
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma2ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attn_logit_softcapping": 50.0,
8
+ "bos_token_id": 2,
9
+ "cache_implementation": "hybrid",
10
+ "eos_token_id": 1,
11
+ "final_logit_softcapping": 30.0,
12
+ "head_dim": 256,
13
+ "hidden_act": "gelu_pytorch_tanh",
14
+ "hidden_activation": "gelu_pytorch_tanh",
15
+ "hidden_size": 3584,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 14336,
18
+ "max_position_embeddings": 8192,
19
+ "model_type": "gemma2",
20
+ "num_attention_heads": 16,
21
+ "num_hidden_layers": 42,
22
+ "num_key_value_heads": 8,
23
+ "pad_token_id": 0,
24
+ "quantization": {
25
+ "group_size": 64,
26
+ "bits": 4
27
+ },
28
+ "quantization_config": {
29
+ "group_size": 64,
30
+ "bits": 4
31
+ },
32
+ "query_pre_attn_scalar": 224,
33
+ "rms_norm_eps": 1e-06,
34
+ "rope_theta": 10000.0,
35
+ "sliding_window": 4096,
36
+ "sliding_window_size": 4096,
37
+ "torch_dtype": "bfloat16",
38
+ "transformers_version": "4.45.1",
39
+ "use_cache": false,
40
+ "vocab_size": 256000
41
+ }