shubhrapandit commited on
Commit
bf3e01b
·
1 Parent(s): c06a2c8

Update files to match base model keeping quant_config intact

Browse files
added_tokens.json DELETED
@@ -1,16 +0,0 @@
1
- {
2
- "<|box_end|>": 151649,
3
- "<|box_start|>": 151648,
4
- "<|endoftext|>": 151643,
5
- "<|im_end|>": 151645,
6
- "<|im_start|>": 151644,
7
- "<|image_pad|>": 151655,
8
- "<|object_ref_end|>": 151647,
9
- "<|object_ref_start|>": 151646,
10
- "<|quad_end|>": 151651,
11
- "<|quad_start|>": 151650,
12
- "<|video_pad|>": 151656,
13
- "<|vision_end|>": 151653,
14
- "<|vision_pad|>": 151654,
15
- "<|vision_start|>": 151652
16
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
chat_template.json CHANGED
@@ -1,3 +1,3 @@
1
  {
2
- "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
3
- }
 
1
  {
2
+ "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
3
+ }
config.json CHANGED
@@ -1,14 +1,17 @@
1
  {
2
- "_name_or_path": "Qwen/Qwen2-VL-72B-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": 8192,
11
- "image_token_id": 151655,
12
  "initializer_range": 0.02,
13
  "intermediate_size": 29568,
14
  "max_position_embeddings": 32768,
@@ -17,6 +20,14 @@
17
  "num_attention_heads": 64,
18
  "num_hidden_layers": 80,
19
  "num_key_value_heads": 8,
 
 
 
 
 
 
 
 
20
  "quantization_config": {
21
  "config_groups": {
22
  "group_0": {
@@ -189,32 +200,25 @@
189
  "quant_method": "compressed-tensors",
190
  "quantization_status": "compressed"
191
  },
192
- "rms_norm_eps": 1e-06,
 
 
 
 
 
 
 
 
 
 
 
193
  "rope_scaling": {
 
194
  "mrope_section": [
195
  16,
196
  24,
197
  24
198
- ],
199
- "rope_type": "default",
200
- "type": "default"
201
- },
202
- "rope_theta": 1000000.0,
203
- "sliding_window": 32768,
204
- "tie_word_embeddings": false,
205
- "torch_dtype": "bfloat16",
206
- "transformers_version": "4.47.1",
207
- "use_cache": true,
208
- "use_sliding_window": false,
209
- "video_token_id": 151656,
210
- "vision_config": {
211
- "hidden_size": 8192,
212
- "in_chans": 3,
213
- "model_type": "qwen2_vl",
214
- "spatial_patch_size": 14
215
  },
216
- "vision_end_token_id": 151653,
217
- "vision_start_token_id": 151652,
218
- "vision_token_id": 151654,
219
  "vocab_size": 152064
220
- }
 
1
  {
 
2
  "architectures": [
3
  "Qwen2VLForConditionalGeneration"
4
  ],
5
  "attention_dropout": 0.0,
6
  "bos_token_id": 151643,
7
  "eos_token_id": 151645,
8
+ "vision_start_token_id": 151652,
9
+ "vision_end_token_id": 151653,
10
+ "vision_token_id": 151654,
11
+ "image_token_id": 151655,
12
+ "video_token_id": 151656,
13
  "hidden_act": "silu",
14
  "hidden_size": 8192,
 
15
  "initializer_range": 0.02,
16
  "intermediate_size": 29568,
17
  "max_position_embeddings": 32768,
 
20
  "num_attention_heads": 64,
21
  "num_hidden_layers": 80,
22
  "num_key_value_heads": 8,
23
+ "rms_norm_eps": 1e-06,
24
+ "rope_theta": 1000000.0,
25
+ "sliding_window": 32768,
26
+ "tie_word_embeddings": false,
27
+ "torch_dtype": "bfloat16",
28
+ "transformers_version": "4.41.2",
29
+ "use_cache": true,
30
+ "use_sliding_window": false,
31
  "quantization_config": {
32
  "config_groups": {
33
  "group_0": {
 
200
  "quant_method": "compressed-tensors",
201
  "quantization_status": "compressed"
202
  },
203
+ "vision_config": {
204
+ "depth": 32,
205
+ "embed_dim": 1280,
206
+ "mlp_ratio": 4,
207
+ "num_heads": 16,
208
+ "in_chans": 3,
209
+ "hidden_size": 8192,
210
+ "patch_size": 14,
211
+ "spatial_merge_size": 2,
212
+ "spatial_patch_size": 14,
213
+ "temporal_patch_size": 2
214
+ },
215
  "rope_scaling": {
216
+ "type": "mrope",
217
  "mrope_section": [
218
  16,
219
  24,
220
  24
221
+ ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
222
  },
 
 
 
223
  "vocab_size": 152064
224
+ }
generation_config.json CHANGED
@@ -1,13 +1,14 @@
1
  {
2
  "bos_token_id": 151643,
 
3
  "do_sample": true,
4
  "eos_token_id": [
5
  151645,
6
  151643
7
  ],
8
- "pad_token_id": 151643,
9
  "repetition_penalty": 1.05,
10
- "top_k": 1,
11
  "top_p": 0.001,
12
- "transformers_version": "4.48.1"
 
13
  }
 
1
  {
2
  "bos_token_id": 151643,
3
+ "pad_token_id": 151643,
4
  "do_sample": true,
5
  "eos_token_id": [
6
  151645,
7
  151643
8
  ],
 
9
  "repetition_penalty": 1.05,
10
+ "temperature": 1,
11
  "top_p": 0.001,
12
+ "top_k": 1,
13
+ "transformers_version": "4.37.0"
14
  }
merges.txt CHANGED
@@ -1,4 +1,3 @@
1
- #version: 0.2
2
  Ġ Ġ
3
  ĠĠ ĠĠ
4
  i n
 
 
1
  Ġ Ġ
2
  ĠĠ ĠĠ
3
  i n
preprocessor_config.json CHANGED
@@ -1,29 +1,19 @@
1
  {
2
- "do_convert_rgb": true,
3
- "do_normalize": true,
4
- "do_rescale": true,
5
- "do_resize": true,
 
6
  "image_mean": [
7
  0.48145466,
8
  0.4578275,
9
  0.40821073
10
  ],
11
- "image_processor_type": "Qwen2VLImageProcessor",
12
  "image_std": [
13
  0.26862954,
14
  0.26130258,
15
  0.27577711
16
  ],
17
- "max_pixels": 12845056,
18
- "merge_size": 2,
19
- "min_pixels": 3136,
20
- "patch_size": 14,
21
- "processor_class": "Qwen2VLProcessor",
22
- "resample": 3,
23
- "rescale_factor": 0.00392156862745098,
24
- "size": {
25
- "max_pixels": 12845056,
26
- "min_pixels": 3136
27
- },
28
- "temporal_patch_size": 2
29
- }
 
1
  {
2
+ "min_pixels": 3136,
3
+ "max_pixels": 12845056,
4
+ "patch_size": 14,
5
+ "temporal_patch_size": 2,
6
+ "merge_size": 2,
7
  "image_mean": [
8
  0.48145466,
9
  0.4578275,
10
  0.40821073
11
  ],
 
12
  "image_std": [
13
  0.26862954,
14
  0.26130258,
15
  0.27577711
16
  ],
17
+ "image_processor_type": "Qwen2VLImageProcessor",
18
+ "processor_class": "Qwen2VLProcessor"
19
+ }
 
 
 
 
 
 
 
 
 
 
special_tokens_map.json DELETED
@@ -1,31 +0,0 @@
1
- {
2
- "additional_special_tokens": [
3
- "<|im_start|>",
4
- "<|im_end|>",
5
- "<|object_ref_start|>",
6
- "<|object_ref_end|>",
7
- "<|box_start|>",
8
- "<|box_end|>",
9
- "<|quad_start|>",
10
- "<|quad_end|>",
11
- "<|vision_start|>",
12
- "<|vision_end|>",
13
- "<|vision_pad|>",
14
- "<|image_pad|>",
15
- "<|video_pad|>"
16
- ],
17
- "eos_token": {
18
- "content": "<|im_end|>",
19
- "lstrip": false,
20
- "normalized": false,
21
- "rstrip": false,
22
- "single_word": false
23
- },
24
- "pad_token": {
25
- "content": "<|endoftext|>",
26
- "lstrip": false,
27
- "normalized": false,
28
- "rstrip": false,
29
- "single_word": false
30
- }
31
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
tokenizer.json CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:268c96d8611372617e20f95ad1eedab8c90f816f1f99ce13355e41b7700f7111
3
- size 11420470
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb63a0a23eef3d5b01063a9880a1925a65aaf4d1591d519910ee3527852950a0
3
+ size 7029741
tokenizer_config.json CHANGED
@@ -114,31 +114,15 @@
114
  "special": true
115
  }
116
  },
117
- "additional_special_tokens": [
118
- "<|im_start|>",
119
- "<|im_end|>",
120
- "<|object_ref_start|>",
121
- "<|object_ref_end|>",
122
- "<|box_start|>",
123
- "<|box_end|>",
124
- "<|quad_start|>",
125
- "<|quad_end|>",
126
- "<|vision_start|>",
127
- "<|vision_end|>",
128
- "<|vision_pad|>",
129
- "<|image_pad|>",
130
- "<|video_pad|>"
131
- ],
132
  "bos_token": null,
133
  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
134
  "clean_up_tokenization_spaces": false,
135
  "eos_token": "<|im_end|>",
 
136
  "errors": "replace",
137
- "extra_special_tokens": {},
138
  "model_max_length": 32768,
139
  "pad_token": "<|endoftext|>",
140
- "padding_side": "left",
141
- "processor_class": "Qwen2VLProcessor",
142
  "split_special_tokens": false,
143
  "tokenizer_class": "Qwen2Tokenizer",
144
  "unk_token": null
 
114
  "special": true
115
  }
116
  },
117
+ "additional_special_tokens": ["<|im_start|>", "<|im_end|>", "<|object_ref_start|>","<|object_ref_end|>","<|box_start|>","<|box_end|>","<|quad_start|>","<|quad_end|>","<|vision_start|>","<|vision_end|>","<|vision_pad|>","<|image_pad|>","<|video_pad|>"],
 
 
 
 
 
 
 
 
 
 
 
 
 
 
118
  "bos_token": null,
119
  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
120
  "clean_up_tokenization_spaces": false,
121
  "eos_token": "<|im_end|>",
122
+ "padding_side": "left",
123
  "errors": "replace",
 
124
  "model_max_length": 32768,
125
  "pad_token": "<|endoftext|>",
 
 
126
  "split_special_tokens": false,
127
  "tokenizer_class": "Qwen2Tokenizer",
128
  "unk_token": null