Hugging Face's logo

prithivMLmods
/
Qwen2-VL-OCR-2B-Instruct

Image-Text-to-Text
Transformers
Safetensors
English
qwen2_vl
Math
OCR
Latex
VLM
Plain_Text
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
prithivMLmods commited on
Commit
01e9286
•
1 Parent(s): 10d2cf0

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +49 -50
config.json CHANGED
@@ -1,50 +1,49 @@
1
- {
2
- "_name_or_path": "Qwen/Qwen2-VL-2B-Instruct",
3
- "architectures": [
4
- "Qwen2VLForConditionalGeneration"
5
- ],
6
- "attention_dropout": 0.0,
7
- "bos_token_id": 151643,
8
- "eos_token_id": 151645,
9
- "hidden_act": "silu",
10
- "hidden_size": 1536,
11
- "image_token_id": 151655,
12
- "initializer_range": 0.02,
13
- "intermediate_size": 8960,
14
- "max_position_embeddings": 32768,
15
- "max_window_layers": 28,
16
- "model_type": "qwen2_vl",
17
- "num_attention_heads": 12,
18
- "num_hidden_layers": 28,
19
- "num_key_value_heads": 2,
20
- "pad_token_id": 151654,
21
- "rms_norm_eps": 1e-06,
22
- "rope_scaling": {
23
- "mrope_section": [
24
- 16,
25
- 24,
26
- 24
27
- ],
28
- "rope_type": "default",
29
- "type": "default"
30
- },
31
- "rope_theta": 1000000.0,
32
- "sliding_window": 32768,
33
- "tie_word_embeddings": true,
34
- "torch_dtype": "bfloat16",
35
- "transformers_version": "4.46.3",
36
- "unsloth_fixed": true,
37
- "use_cache": true,
38
- "use_sliding_window": false,
39
- "video_token_id": 151656,
40
- "vision_config": {
41
- "hidden_size": 1536,
42
- "in_chans": 3,
43
- "model_type": "qwen2_vl",
44
- "spatial_patch_size": 14
45
- },
46
- "vision_end_token_id": 151653,
47
- "vision_start_token_id": 151652,
48
- "vision_token_id": 151654,
49
- "vocab_size": 151936
50
- }
 
1
+ {
2
+ "_name_or_path": "Qwen/Qwen2-VL-2B-Instruct",
3
+ "architectures": [
4
+ "Qwen2VLForConditionalGeneration"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 1536,
11
+ "image_token_id": 151655,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8960,
14
+ "max_position_embeddings": 32768,
15
+ "max_window_layers": 28,
16
+ "model_type": "qwen2_vl",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 28,
19
+ "num_key_value_heads": 2,
20
+ "pad_token_id": 151654,
21
+ "rms_norm_eps": 1e-06,
22
+ "rope_scaling": {
23
+ "mrope_section": [
24
+ 16,
25
+ 24,
26
+ 24
27
+ ],
28
+ "rope_type": "default",
29
+ "type": "default"
30
+ },
31
+ "rope_theta": 1000000.0,
32
+ "sliding_window": 32768,
33
+ "tie_word_embeddings": true,
34
+ "torch_dtype": "bfloat16",
35
+ "transformers_version": "4.46.3",
36
+ "use_cache": true,
37
+ "use_sliding_window": false,
38
+ "video_token_id": 151656,
39
+ "vision_config": {
40
+ "hidden_size": 1536,
41
+ "in_chans": 3,
42
+ "model_type": "qwen2_vl",
43
+ "spatial_patch_size": 14
44
+ },
45
+ "vision_end_token_id": 151653,
46
+ "vision_start_token_id": 151652,
47
+ "vision_token_id": 151654,
48
+ "vocab_size": 151936
49
+ }