Hugging Face's logo

ashun989
/
GlimpsePrune_Qwen2.5-VL-3B-Instruct

Image-Text-to-Text
Transformers
qwen2_5_vl_gp
Model card Files
xet
 Community
1
ashun989 commited on
Commit
f2cf4a0
·
verified ·
1 Parent(s): a36c700

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. config.json +132 -0
  2. new_modules_gp.pt +3 -0
config.json ADDED
@@ -0,0 +1,132 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "anchor_positions": null,
4
+ "architectures": [
5
+ "Qwen2_5_VLForConditionalGeneration"
6
+ ],
7
+ "attention_dropout": 0.0,
8
+ "attn_fuse_global": true,
9
+ "attn_fuse_hidden_act": "silu",
10
+ "attn_fuse_num_heads": 4,
11
+ "attn_fuse_size": 256,
12
+ "attn_fuse_type": "AttnFuserV1",
13
+ "bos_token_id": 151643,
14
+ "deep_supervision": false,
15
+ "eos_token_id": 151645,
16
+ "hidden_act": "silu",
17
+ "hidden_size": 2048,
18
+ "image_token_id": 151655,
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 11008,
21
+ "le_dropout_prob": 0.1,
22
+ "le_layers": [
23
+ 0,
24
+ 1,
25
+ 2,
26
+ 3,
27
+ 4,
28
+ 5,
29
+ 6,
30
+ 7,
31
+ 8,
32
+ 9,
33
+ 10,
34
+ 11,
35
+ 12,
36
+ 13,
37
+ 14,
38
+ 15,
39
+ 16,
40
+ 17,
41
+ 18,
42
+ 19,
43
+ 20,
44
+ 21,
45
+ 22,
46
+ 23,
47
+ 24,
48
+ 25,
49
+ 26,
50
+ 27,
51
+ 28,
52
+ 29,
53
+ 30,
54
+ 31,
55
+ 32,
56
+ 33,
57
+ 34,
58
+ 35
59
+ ],
60
+ "le_length": 1,
61
+ "le_norm_type": "rmsnorm",
62
+ "max_position_embeddings": 128000,
63
+ "max_remain_ratio": null,
64
+ "max_window_layers": 70,
65
+ "min_remain_num": null,
66
+ "model_type": "qwen2_5_vl_gp",
67
+ "num_attention_heads": 16,
68
+ "num_hidden_layers": 36,
69
+ "num_key_value_heads": 2,
70
+ "ori_attn_supervision": false,
71
+ "reduce_layer": 23,
72
+ "reduce_threshold": 0.5,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_scaling": {
75
+ "mrope_section": [
76
+ 16,
77
+ 24,
78
+ 24
79
+ ],
80
+ "rope_type": "default",
81
+ "type": "default"
82
+ },
83
+ "rope_theta": 1000000.0,
84
+ "selected_layers": [
85
+ 23
86
+ ],
87
+ "selected_visual_layers": [
88
+ 31,
89
+ 23,
90
+ 15,
91
+ 7
92
+ ],
93
+ "sliding_window": 32768,
94
+ "tie_word_embeddings": true,
95
+ "torch_dtype": "bfloat16",
96
+ "transformers_version": "4.51.3",
97
+ "use_attention_logits": true,
98
+ "use_cache": true,
99
+ "use_ref_masks": false,
100
+ "use_sliding_window": false,
101
+ "use_zero_masks": false,
102
+ "video_token_id": 151656,
103
+ "vision_config": {
104
+ "depth": 32,
105
+ "fullatt_block_indexes": [
106
+ 7,
107
+ 15,
108
+ 23,
109
+ 31
110
+ ],
111
+ "hidden_act": "silu",
112
+ "hidden_size": 1280,
113
+ "in_channels": 3,
114
+ "in_chans": 3,
115
+ "intermediate_size": 3420,
116
+ "model_type": "qwen2_5_vl",
117
+ "num_heads": 16,
118
+ "out_hidden_size": 2048,
119
+ "patch_size": 14,
120
+ "spatial_merge_size": 2,
121
+ "spatial_patch_size": 14,
122
+ "temporal_patch_size": 2,
123
+ "tokens_per_second": 2,
124
+ "torch_dtype": "bfloat16",
125
+ "window_size": 112
126
+ },
127
+ "vision_end_token_id": 151653,
128
+ "vision_start_token_id": 151652,
129
+ "vision_token_id": 151654,
130
+ "visual_cond_size": 512,
131
+ "vocab_size": 151936
132
+ }
new_modules_gp.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fc73ffefef616dee571e71849f678658f7e998afc6d21339b783a38f64c205e
3
+ size 27470349