zeng981 commited on
Commit
2edb169
·
verified ·
1 Parent(s): 040f8b1

Upload config.json

Browse files
Files changed (1) hide show
  1. config.json +66 -0
config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "D:\\models\\gemma-2-9b",
3
+ "add_bias_linear": false,
4
+ "add_qkv_bias": true,
5
+ "apply_query_key_layer_scaling": true,
6
+ "apply_residual_connection_post_layernorm": false,
7
+ "architectures": [
8
+ "GemmaForCausalLM"
9
+ ],
10
+ "attention_dropout": 0.0,
11
+ "attention_softmax_in_fp32": true,
12
+ "auto_map": {
13
+ "AutoConfig": "configuration_gemma.GemmaConfig",
14
+ "AutoModel": "modeling_gemma.GemmaForCausalLM",
15
+ "AutoModelForCausalLM": "modeling_gemma.GemmaForCausalLM",
16
+ "AutoModelForSeq2SeqLM": "modeling_gemma.GemmaForCausalLM",
17
+ "AutoModelForSequenceClassification": "modeling_gemma.GemmaForSequenceClassification"
18
+ },
19
+ "bias_dropout_fusion": true,
20
+ "boi_token_id": 1,
21
+ "classifier_dropout": null,
22
+ "eoi_token_id": 2,
23
+ "eos_token_id": [
24
+ 1,
25
+ 2
26
+ ],
27
+ "ffn_hidden_size": 12288,
28
+ "fp32_residual_connection": false,
29
+ "hidden_dropout": 0.0,
30
+ "hidden_size": 3584,
31
+ "kv_channels": 128,
32
+ "layernorm_epsilon": 1.0e-6,
33
+ "model_type": "gemma2",
34
+ "multi_query_attention": true,
35
+ "multi_query_group_num": 2,
36
+ "num_attention_heads": 16,
37
+ "num_layers": 42,
38
+ "original_rope": true,
39
+ "pad_token_id": 0,
40
+ "padded_vocab_size": 256000,
41
+ "post_layer_norm": true,
42
+ "pre_seq_len": null,
43
+ "prefix_projection": false,
44
+ "rmsnorm": true,
45
+ "rope_ratio": 1,
46
+ "seq_length": 8192,
47
+ "tie_word_embeddings": false,
48
+ "torch_dtype": "float32",
49
+ "transformers_version": "4.42.3",
50
+ "use_cache": true,
51
+ "vision_config": {
52
+ "dropout_prob": 0.0,
53
+ "hidden_act": "gelu",
54
+ "hidden_size": 2048,
55
+ "image_size": 1120,
56
+ "in_channels": 3,
57
+ "intermediate_size": 15360,
58
+ "layer_norm_eps": 1.0e-6,
59
+ "num_heads": 16,
60
+ "num_hidden_layers": 63,
61
+ "num_positions": 6401,
62
+ "patch_size": 14,
63
+ "scaling_factor": 8
64
+ },
65
+ "vocab_size": 256000
66
+ }