waiyiaisg commited on
Commit
1ed947d
·
verified ·
1 Parent(s): afd2c2f

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +3 -83
config.json CHANGED
@@ -13,111 +13,31 @@
13
  "mm_tokens_per_image": 256,
14
  "model_type": "gemma3",
15
  "text_config": {
16
- "_sliding_window_pattern": 6,
17
- "attention_bias": false,
18
- "attention_dropout": 0.0,
19
- "attn_logit_softcapping": null,
20
- "final_logit_softcapping": null,
21
  "head_dim": 128,
22
- "hidden_activation": "gelu_pytorch_tanh",
23
  "hidden_size": 5376,
24
- "initializer_range": 0.02,
25
  "intermediate_size": 21504,
26
- "layer_types": [
27
- "sliding_attention",
28
- "sliding_attention",
29
- "sliding_attention",
30
- "sliding_attention",
31
- "sliding_attention",
32
- "full_attention",
33
- "sliding_attention",
34
- "sliding_attention",
35
- "sliding_attention",
36
- "sliding_attention",
37
- "sliding_attention",
38
- "full_attention",
39
- "sliding_attention",
40
- "sliding_attention",
41
- "sliding_attention",
42
- "sliding_attention",
43
- "sliding_attention",
44
- "full_attention",
45
- "sliding_attention",
46
- "sliding_attention",
47
- "sliding_attention",
48
- "sliding_attention",
49
- "sliding_attention",
50
- "full_attention",
51
- "sliding_attention",
52
- "sliding_attention",
53
- "sliding_attention",
54
- "sliding_attention",
55
- "sliding_attention",
56
- "full_attention",
57
- "sliding_attention",
58
- "sliding_attention",
59
- "sliding_attention",
60
- "sliding_attention",
61
- "sliding_attention",
62
- "full_attention",
63
- "sliding_attention",
64
- "sliding_attention",
65
- "sliding_attention",
66
- "sliding_attention",
67
- "sliding_attention",
68
- "full_attention",
69
- "sliding_attention",
70
- "sliding_attention",
71
- "sliding_attention",
72
- "sliding_attention",
73
- "sliding_attention",
74
- "full_attention",
75
- "sliding_attention",
76
- "sliding_attention",
77
- "sliding_attention",
78
- "sliding_attention",
79
- "sliding_attention",
80
- "full_attention",
81
- "sliding_attention",
82
- "sliding_attention",
83
- "sliding_attention",
84
- "sliding_attention",
85
- "sliding_attention",
86
- "full_attention",
87
- "sliding_attention",
88
- "sliding_attention"
89
- ],
90
- "max_position_embeddings": 131072,
91
  "model_type": "gemma3_text",
92
  "num_attention_heads": 32,
93
  "num_hidden_layers": 62,
94
  "num_key_value_heads": 16,
95
  "query_pre_attn_scalar": 168,
96
- "rms_norm_eps": 1e-06,
97
- "rope_local_base_freq": 10000.0,
98
  "rope_scaling": {
99
  "factor": 8.0,
100
  "rope_type": "linear"
101
  },
102
- "rope_theta": 1000000.0,
103
- "sliding_window": 1024,
104
- "use_cache": true,
105
- "vocab_size": 262208
106
  },
107
  "torch_dtype": "bfloat16",
108
- "transformers_version": "4.53.2",
109
  "vision_config": {
110
- "attention_dropout": 0.0,
111
- "hidden_act": "gelu_pytorch_tanh",
112
  "hidden_size": 1152,
113
  "image_size": 896,
114
  "intermediate_size": 4304,
115
- "layer_norm_eps": 1e-06,
116
  "model_type": "siglip_vision_model",
117
  "num_attention_heads": 16,
118
- "num_channels": 3,
119
  "num_hidden_layers": 27,
120
  "patch_size": 14,
121
  "vision_use_head": false
122
  }
123
  }
 
 
13
  "mm_tokens_per_image": 256,
14
  "model_type": "gemma3",
15
  "text_config": {
 
 
 
 
 
16
  "head_dim": 128,
 
17
  "hidden_size": 5376,
 
18
  "intermediate_size": 21504,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
19
  "model_type": "gemma3_text",
20
  "num_attention_heads": 32,
21
  "num_hidden_layers": 62,
22
  "num_key_value_heads": 16,
23
  "query_pre_attn_scalar": 168,
 
 
24
  "rope_scaling": {
25
  "factor": 8.0,
26
  "rope_type": "linear"
27
  },
28
+ "sliding_window": 1024
 
 
 
29
  },
30
  "torch_dtype": "bfloat16",
31
+ "transformers_version": "4.50.0.dev0",
32
  "vision_config": {
 
 
33
  "hidden_size": 1152,
34
  "image_size": 896,
35
  "intermediate_size": 4304,
 
36
  "model_type": "siglip_vision_model",
37
  "num_attention_heads": 16,
 
38
  "num_hidden_layers": 27,
39
  "patch_size": 14,
40
  "vision_use_head": false
41
  }
42
  }
43
+