Hugging Face's logo

yifanzhang114
/
MM-RLHF-Reward-7B-llava-ov-qwen

Image-Text-to-Text
Transformers
Safetensors
llava_qwen
text-generation
conversational
Inference Endpoints
Model card Files Community
1
yifanzhang114 commited on
Commit
c761268
·
verified ·
1 Parent(s): 5cd958e

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +121 -0
config.json ADDED
@@ -0,0 +1,121 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "lmms-lab/llava-onevision-qwen2-7b-ov",
3
+ "architectures": [
4
+ "LlavaQwenForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 3584,
11
+ "image_aspect_ratio": "anyres_max_4",
12
+ "image_crop_resolution": 384,
13
+ "image_grid_pinpoints": [
14
+ [
15
+ 384,
16
+ 384
17
+ ],
18
+ [
19
+ 384,
20
+ 768
21
+ ],
22
+ [
23
+ 384,
24
+ 1152
25
+ ],
26
+ [
27
+ 384,
28
+ 1536
29
+ ],
30
+ [
31
+ 768,
32
+ 384
33
+ ],
34
+ [
35
+ 768,
36
+ 768
37
+ ],
38
+ [
39
+ 768,
40
+ 1152
41
+ ],
42
+ [
43
+ 768,
44
+ 1536
45
+ ],
46
+ [
47
+ 1152,
48
+ 384
49
+ ],
50
+ [
51
+ 1152,
52
+ 768
53
+ ],
54
+ [
55
+ 1152,
56
+ 1152
57
+ ],
58
+ [
59
+ 1152,
60
+ 1536
61
+ ],
62
+ [
63
+ 1536,
64
+ 384
65
+ ],
66
+ [
67
+ 1536,
68
+ 768
69
+ ],
70
+ [
71
+ 1536,
72
+ 1152
73
+ ],
74
+ [
75
+ 1536,
76
+ 1536
77
+ ]
78
+ ],
79
+ "image_split_resolution": 384,
80
+ "image_token_index": 151646,
81
+ "initializer_range": 0.02,
82
+ "intermediate_size": 18944,
83
+ "is_rm": true,
84
+ "max_position_embeddings": 32768,
85
+ "max_window_layers": 28,
86
+ "mm_hidden_size": 1152,
87
+ "mm_newline_position": "one_token",
88
+ "mm_patch_merge_type": "spatial_unpad",
89
+ "mm_projector_lr": null,
90
+ "mm_projector_type": "mlp2x_gelu",
91
+ "mm_resampler_type": null,
92
+ "mm_spatial_pool_mode": "bilinear",
93
+ "mm_tunable_parts": "mm_vision_tower,mm_mlp_adapter,mm_language_model",
94
+ "mm_use_im_patch_token": false,
95
+ "mm_use_im_start_end": false,
96
+ "mm_vision_select_feature": "patch",
97
+ "mm_vision_select_layer": -2,
98
+ "mm_vision_tower": "google/siglip-so400m-patch14-384",
99
+ "mm_vision_tower_lr": null,
100
+ "model_type": "llava_qwen",
101
+ "num_attention_heads": 28,
102
+ "num_hidden_layers": 28,
103
+ "num_key_value_heads": 4,
104
+ "pad_token_id": 151643,
105
+ "pos_skipping_range": 4096,
106
+ "rms_norm_eps": 1e-06,
107
+ "rope_scaling": null,
108
+ "rope_theta": 1000000.0,
109
+ "sliding_window": 131072,
110
+ "tie_word_embeddings": false,
111
+ "tokenizer_model_max_length": 20480,
112
+ "tokenizer_padding_side": "right",
113
+ "torch_dtype": "bfloat16",
114
+ "transformers_version": "4.40.0.dev0",
115
+ "use_cache": true,
116
+ "use_mm_proj": true,
117
+ "use_pos_skipping": false,
118
+ "use_sliding_window": false,
119
+ "vision_tower_pretrained": null,
120
+ "vocab_size": 152064
121
+ }