Ephraimmm commited on
Commit
2919d76
·
verified ·
1 Parent(s): 7261fe9

Delete config.json

Browse files
Files changed (1) hide show
  1. config.json +0 -143
config.json DELETED
@@ -1,143 +0,0 @@
1
- {
2
- "architectures": [
3
- "Gemma3ForConditionalGeneration"
4
- ],
5
- "boi_token_index": 255999,
6
- "bos_token_id": 2,
7
- "eoi_token_index": 256000,
8
- "eos_token_id": 106,
9
- "image_token_index": 262144,
10
- "initializer_range": 0.02,
11
- "mm_tokens_per_image": 256,
12
- "model_type": "gemma3",
13
- "pad_token_id": 0,
14
- "quantization_config": {
15
- "_load_in_4bit": true,
16
- "_load_in_8bit": false,
17
- "bnb_4bit_compute_dtype": "bfloat16",
18
- "bnb_4bit_quant_storage": "uint8",
19
- "bnb_4bit_quant_type": "nf4",
20
- "bnb_4bit_use_double_quant": true,
21
- "llm_int8_enable_fp32_cpu_offload": false,
22
- "llm_int8_has_fp16_weight": false,
23
- "llm_int8_skip_modules": [
24
- "lm_head",
25
- "multi_modal_projector",
26
- "merger",
27
- "modality_projection",
28
- "language_model.model.layers.1.mlp",
29
- "language_model.model.layers.4.mlp",
30
- "language_model.model.layers.2.mlp",
31
- "language_model.model.layers.5.mlp",
32
- "vision_tower.vision_model.encoder.layers.22.self_attn",
33
- "language_model.model.layers.3.mlp",
34
- "vision_tower.vision_model.encoder.layers.26.self_attn",
35
- "language_model.model.layers.0.mlp",
36
- "vision_tower.vision_model.encoder.layers.19.self_attn",
37
- "vision_tower.vision_model.encoder.layers.25.self_attn",
38
- "vision_tower.vision_model.encoder.layers.23.self_attn",
39
- "vision_tower.vision_model.encoder.layers.25.mlp",
40
- "vision_tower.vision_model.encoder.layers.20.self_attn",
41
- "vision_tower.vision_model.encoder.layers.24.self_attn",
42
- "language_model.model.layers.2.self_attn",
43
- "vision_tower.vision_model.encoder.layers.21.self_attn",
44
- "vision_tower.vision_model.encoder.layers.18.self_attn",
45
- "vision_tower.vision_model.encoder.layers.19.mlp",
46
- "vision_tower.vision_model.encoder.layers.20.mlp",
47
- "vision_tower.vision_model.encoder.layers.21.mlp",
48
- "vision_tower.vision_model.encoder.layers.15.mlp",
49
- "vision_tower.vision_model.encoder.layers.22.mlp",
50
- "vision_tower.vision_model.encoder.layers.23.mlp",
51
- "vision_tower.vision_model.encoder.layers.17.mlp",
52
- "vision_tower.vision_model.encoder.layers.18.mlp",
53
- "vision_tower.vision_model.encoder.layers.24.mlp",
54
- "vision_tower.vision_model.encoder.layers.16.self_attn",
55
- "vision_tower.vision_model.encoder.layers.16.mlp",
56
- "vision_tower.vision_model.encoder.layers.17.self_attn",
57
- "vision_tower.vision_model.encoder.layers.13.mlp",
58
- "vision_tower.vision_model.encoder.layers.14.mlp",
59
- "vision_tower.vision_model.encoder.layers.15.self_attn",
60
- "vision_tower.vision_model.encoder.layers.12.mlp",
61
- "vision_tower.vision_model.encoder.layers.14.self_attn",
62
- "vision_tower.vision_model.encoder.layers.9.mlp",
63
- "vision_tower.vision_model.encoder.layers.5.mlp",
64
- "vision_tower.vision_model.encoder.layers.10.self_attn",
65
- "vision_tower.vision_model.encoder.layers.7.mlp",
66
- "vision_tower.vision_model.encoder.layers.9.self_attn",
67
- "vision_tower.vision_model.encoder.layers.8.mlp",
68
- "vision_tower.vision_model.encoder.layers.10.mlp",
69
- "vision_tower.vision_model.encoder.layers.13.self_attn",
70
- "vision_tower.vision_model.encoder.layers.11.mlp",
71
- "vision_tower.vision_model.encoder.layers.11.self_attn",
72
- "vision_tower.vision_model.encoder.layers.12.self_attn",
73
- "vision_tower.vision_model.encoder.layers.3.mlp",
74
- "vision_tower.vision_model.encoder.layers.8.self_attn",
75
- "vision_tower.vision_model.encoder.layers.4.self_attn",
76
- "vision_tower.vision_model.encoder.layers.6.mlp",
77
- "vision_tower.vision_model.encoder.layers.4.mlp",
78
- "vision_tower.vision_model.encoder.layers.2.mlp",
79
- "vision_tower.vision_model.encoder.layers.6.self_attn",
80
- "vision_tower.vision_model.encoder.layers.7.self_attn",
81
- "vision_tower.vision_model.encoder.layers.1.self_attn",
82
- "vision_tower.vision_model.encoder.layers.1.mlp",
83
- "vision_tower.vision_model.encoder.layers.3.self_attn",
84
- "vision_tower.vision_model.encoder.layers.5.self_attn",
85
- "vision_tower.vision_model.encoder.layers.0.self_attn",
86
- "vision_tower.vision_model.encoder.layers.0.mlp",
87
- "vision_tower.vision_model.encoder.layers.2.self_attn",
88
- "vision_tower.vision_model.encoder.layers.26.mlp"
89
- ],
90
- "llm_int8_threshold": 6.0,
91
- "load_in_4bit": true,
92
- "load_in_8bit": false,
93
- "quant_method": "bitsandbytes"
94
- },
95
- "text_config": {
96
- "attention_bias": false,
97
- "attention_dropout": 0.0,
98
- "attn_logit_softcapping": null,
99
- "cache_implementation": "hybrid",
100
- "final_logit_softcapping": null,
101
- "head_dim": 256,
102
- "hidden_activation": "gelu_pytorch_tanh",
103
- "hidden_size": 2560,
104
- "initializer_range": 0.02,
105
- "intermediate_size": 10240,
106
- "max_position_embeddings": 131072,
107
- "model_type": "gemma3_text",
108
- "num_attention_heads": 8,
109
- "num_hidden_layers": 34,
110
- "num_key_value_heads": 4,
111
- "query_pre_attn_scalar": 256,
112
- "rms_norm_eps": 1e-06,
113
- "rope_local_base_freq": 10000.0,
114
- "rope_scaling": {
115
- "factor": 8.0,
116
- "rope_type": "linear"
117
- },
118
- "rope_theta": 1000000.0,
119
- "sliding_window": 1024,
120
- "sliding_window_pattern": 6,
121
- "torch_dtype": "bfloat16",
122
- "use_cache": true,
123
- "vocab_size": 262208
124
- },
125
- "torch_dtype": "bfloat16",
126
- "transformers_version": "4.51.3",
127
- "unsloth_fixed": true,
128
- "vision_config": {
129
- "attention_dropout": 0.0,
130
- "hidden_act": "gelu_pytorch_tanh",
131
- "hidden_size": 1152,
132
- "image_size": 896,
133
- "intermediate_size": 4304,
134
- "layer_norm_eps": 1e-06,
135
- "model_type": "siglip_vision_model",
136
- "num_attention_heads": 16,
137
- "num_channels": 3,
138
- "num_hidden_layers": 27,
139
- "patch_size": 14,
140
- "torch_dtype": "bfloat16",
141
- "vision_use_head": false
142
- }
143
- }