jayanth7111 commited on
Commit
e325aaa
·
verified ·
1 Parent(s): 63f63cc

Upload DriveSense-VLM artifacts

Browse files
Files changed (1) hide show
  1. config.json +124 -0
config.json ADDED
@@ -0,0 +1,124 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen2_5_VLForConditionalGeneration"
4
+ ],
5
+ "dtype": "bfloat16",
6
+ "image_token_id": 151655,
7
+ "model_type": "qwen2_5_vl",
8
+ "quantization_config": {
9
+ "_load_in_4bit": true,
10
+ "_load_in_8bit": false,
11
+ "bnb_4bit_compute_dtype": "bfloat16",
12
+ "bnb_4bit_quant_storage": "uint8",
13
+ "bnb_4bit_quant_type": "nf4",
14
+ "bnb_4bit_use_double_quant": true,
15
+ "llm_int8_enable_fp32_cpu_offload": false,
16
+ "llm_int8_has_fp16_weight": false,
17
+ "llm_int8_skip_modules": null,
18
+ "llm_int8_threshold": 6.0,
19
+ "load_in_4bit": true,
20
+ "load_in_8bit": false,
21
+ "quant_method": "bitsandbytes"
22
+ },
23
+ "text_config": {
24
+ "attention_dropout": 0.0,
25
+ "bos_token_id": 151643,
26
+ "dtype": "bfloat16",
27
+ "eos_token_id": 151645,
28
+ "hidden_act": "silu",
29
+ "hidden_size": 2048,
30
+ "initializer_range": 0.02,
31
+ "intermediate_size": 11008,
32
+ "layer_types": [
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention",
53
+ "full_attention",
54
+ "full_attention",
55
+ "full_attention",
56
+ "full_attention",
57
+ "full_attention",
58
+ "full_attention",
59
+ "full_attention",
60
+ "full_attention",
61
+ "full_attention",
62
+ "full_attention",
63
+ "full_attention",
64
+ "full_attention",
65
+ "full_attention",
66
+ "full_attention",
67
+ "full_attention",
68
+ "full_attention"
69
+ ],
70
+ "max_position_embeddings": 128000,
71
+ "max_window_layers": 70,
72
+ "model_type": "qwen2_5_vl_text",
73
+ "num_attention_heads": 16,
74
+ "num_hidden_layers": 36,
75
+ "num_key_value_heads": 2,
76
+ "pad_token_id": null,
77
+ "rms_norm_eps": 1e-06,
78
+ "rope_parameters": {
79
+ "mrope_section": [
80
+ 16,
81
+ 24,
82
+ 24
83
+ ],
84
+ "rope_theta": 1000000.0,
85
+ "rope_type": "default",
86
+ "type": "mrope"
87
+ },
88
+ "sliding_window": null,
89
+ "use_cache": true,
90
+ "use_sliding_window": false,
91
+ "vocab_size": 151936
92
+ },
93
+ "tie_word_embeddings": true,
94
+ "transformers_version": "5.0.0",
95
+ "video_token_id": 151656,
96
+ "vision_config": {
97
+ "depth": 32,
98
+ "dtype": "bfloat16",
99
+ "fullatt_block_indexes": [
100
+ 7,
101
+ 15,
102
+ 23,
103
+ 31
104
+ ],
105
+ "hidden_act": "silu",
106
+ "hidden_size": 1280,
107
+ "in_channels": 3,
108
+ "in_chans": 3,
109
+ "initializer_range": 0.02,
110
+ "intermediate_size": 3420,
111
+ "model_type": "qwen2_5_vl",
112
+ "num_heads": 16,
113
+ "out_hidden_size": 2048,
114
+ "patch_size": 14,
115
+ "spatial_merge_size": 2,
116
+ "spatial_patch_size": 14,
117
+ "temporal_patch_size": 2,
118
+ "tokens_per_second": 2,
119
+ "window_size": 112
120
+ },
121
+ "vision_end_token_id": 151653,
122
+ "vision_start_token_id": 151652,
123
+ "vision_token_id": 151654
124
+ }