yujingfeng commited on
Commit
90cb834
·
verified ·
1 Parent(s): 730ffdb

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +2 -75
config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "tokenizer_class": "Qwen2_5_VLTokenizer",
3
  "architectures": [
4
  "Qwen2_5_VLForCausalLM"
5
  ],
@@ -18,89 +17,17 @@
18
  "num_hidden_layers": 28,
19
  "num_key_value_heads": 4,
20
  "rms_norm_eps": 1e-06,
21
- "rope_scaling": {
22
- "mrope_section": [
23
- 16,
24
- 24,
25
- 24
26
- ],
27
- "rope_type": "default",
28
- "type": "default"
29
- },
30
  "rope_theta": 1000000.0,
31
  "sliding_window": 32768,
32
- "text_config": {
33
- "architectures": [
34
- "Qwen2_5_VLForConditionalGeneration"
35
- ],
36
- "attention_dropout": 0.0,
37
- "bos_token_id": 151643,
38
- "eos_token_id": 151645,
39
- "hidden_act": "silu",
40
- "hidden_size": 3584,
41
- "image_token_id": null,
42
- "initializer_range": 0.02,
43
- "intermediate_size": 18944,
44
- "max_position_embeddings": 128000,
45
- "max_window_layers": 28,
46
- "model_type": "qwen2_5_vl_text",
47
- "num_attention_heads": 28,
48
- "num_hidden_layers": 28,
49
- "num_key_value_heads": 4,
50
- "rms_norm_eps": 1e-06,
51
- "rope_scaling": {
52
- "mrope_section": [
53
- 16,
54
- 24,
55
- 24
56
- ],
57
- "rope_type": "default",
58
- "type": "default"
59
- },
60
- "rope_theta": 1000000.0,
61
- "sliding_window": 32768,
62
- "torch_dtype": "bfloat16",
63
- "use_cache": true,
64
- "use_sliding_window": false,
65
- "video_token_id": null,
66
- "vision_end_token_id": 151653,
67
- "vision_start_token_id": 151652,
68
- "vision_token_id": 151654,
69
- "vocab_size": 152064
70
- },
71
- "tie_word_embeddings": false,
72
  "torch_dtype": "bfloat16",
73
  "transformers_version": "4.52.4",
74
  "use_cache": true,
75
  "use_sliding_window": false,
76
  "video_token_id": 151656,
77
- "vision_config": {
78
- "depth": 32,
79
- "fullatt_block_indexes": [
80
- 7,
81
- 15,
82
- 23,
83
- 31
84
- ],
85
- "hidden_act": "silu",
86
- "hidden_size": 1280,
87
- "in_channels": 3,
88
- "in_chans": 3,
89
- "initializer_range": 0.02,
90
- "intermediate_size": 3420,
91
- "model_type": "qwen2_5_vl",
92
- "num_heads": 16,
93
- "out_hidden_size": 3584,
94
- "patch_size": 14,
95
- "spatial_merge_size": 2,
96
- "spatial_patch_size": 14,
97
- "temporal_patch_size": 2,
98
- "tokens_per_second": 2,
99
- "torch_dtype": "bfloat16",
100
- "window_size": 112
101
- },
102
  "vision_end_token_id": 151653,
103
  "vision_start_token_id": 151652,
104
  "vision_token_id": 151654,
105
  "vocab_size": 152064
106
  }
 
 
1
  {
 
2
  "architectures": [
3
  "Qwen2_5_VLForCausalLM"
4
  ],
 
17
  "num_hidden_layers": 28,
18
  "num_key_value_heads": 4,
19
  "rms_norm_eps": 1e-06,
 
 
 
 
 
 
 
 
 
20
  "rope_theta": 1000000.0,
21
  "sliding_window": 32768,
22
+ "tokenizer_class": "Qwen2_5_VLTokenizer",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
23
  "torch_dtype": "bfloat16",
24
  "transformers_version": "4.52.4",
25
  "use_cache": true,
26
  "use_sliding_window": false,
27
  "video_token_id": 151656,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
28
  "vision_end_token_id": 151653,
29
  "vision_start_token_id": 151652,
30
  "vision_token_id": 151654,
31
  "vocab_size": 152064
32
  }
33
+