VisoLearn ahmedheakl commited on
Commit
1ccd57b
·
verified ·
0 Parent(s):

Duplicate from MBZUAI/AIN

Browse files

Co-authored-by: Ahmed Heakl <[email protected]>

.gitattributes ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz filter=lfs diff=lfs merge=lfs -text
33
+ *.zip filter=lfs diff=lfs merge=lfs -text
34
+ *.zst filter=lfs diff=lfs merge=lfs -text
35
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ assets_hf/ain_can_see.png filter=lfs diff=lfs merge=lfs -text
38
+ assets_hf/AIN.png filter=lfs diff=lfs merge=lfs -text
39
+ assets_hf/Eval_CAMEL.png filter=lfs diff=lfs merge=lfs -text
40
+ assets_hf/qualitative.png filter=lfs diff=lfs merge=lfs -text
41
+ assets_hf/radar_chart.png filter=lfs diff=lfs merge=lfs -text
42
+ assets_hf/toxicity.png filter=lfs diff=lfs merge=lfs -text
43
+ assets_hf/verify_pipeline.png filter=lfs diff=lfs merge=lfs -text
44
+ assets_hf/demo_image.jpeg filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,622 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ language:
4
+ - en
5
+ - ar
6
+ base_model:
7
+ - qwen2-VL-7B
8
+ pipeline_tag: image-text-to-text
9
+ tags:
10
+ - LMM
11
+ - Arabic
12
+ - OCR
13
+ ---
14
+
15
+
16
+ <div style="display: flex; align-items: center;">
17
+ <img src="assets_hf/AIN.png" width="10%" alt="logo" style="margin-right: 10px;" />
18
+ <h1 style="margin: 0; font-size: 28px;";">AIN: The Arabic INclusive Large Multimodal Model</h1>
19
+ </div>
20
+
21
+ [Ahmed Heakl](https://huggingface.co/ahmedheakl) <sup> * </sup> &nbsp;
22
+ [Sara Ghaboura](https://huggingface.co/SLMLAH) <sup> * </sup> &nbsp;
23
+ [Omkar Thawakar](https://omkarthawakar.github.io) &nbsp;
24
+ [Fahad Shahbaz Khan](https://scholar.google.com/citations?hl=en&user=zvaeYnUAAAAJ) &nbsp;
25
+ [Hisham Cholakkal](https://scholar.google.com/citations?hl=en&user=bZ3YBRcAAAAJ) &nbsp;
26
+ [Rao M. Anwer](https://scholar.google.com/citations?hl=en&user=_KlvMVoAAAAJ) &nbsp;
27
+ [Salman Khan](https://scholar.google.com/citations?hl=en&user=M59O9lkAAAAJ)
28
+ <br>
29
+ <em> <sup> *Equal Contribution </sup> </em>
30
+ <br>
31
+ #### **Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI), UAE**
32
+ [![arXiv](https://img.shields.io/badge/arXiv-2502.00094-3399FF)](https://arxiv.org/abs/2502.00094)
33
+ [![Our Page](https://img.shields.io/badge/Visit-Our%20Page-8C7AFF?style=flat)](https://mbzuai-oryx.github.io/AIN/)
34
+ [![Github](https://img.shields.io/badge/Visit-Our%20Github-9BEDB9?style=flat)](https://github.com/mbzuai-oryx/AIN)
35
+ [![GitHub issues](https://img.shields.io/github/issues/mbzuai-oryx/Camel-Bench?color=FFF359&label=issues&style=flat)](https://github.com/mbzuai-oryx/AIN/issues)
36
+ [![GitHub stars](https://img.shields.io/github/stars/mbzuai-oryx/AIN?color=FF6A07&style=flat)](https://github.com/mbzuai-oryx/AIN/stargazers)
37
+ [![GitHub license](https://img.shields.io/github/license/mbzuai-oryx/Camel-Bench?color=FF6666)](https://github.com/mbzuai-oryx/AIN/blob/main/LICENSE)
38
+
39
+ ---
40
+
41
+
42
+ <div class="abstract-container">
43
+ <h2>Abstract</h2>
44
+ <div class="abstract-content">
45
+ <p>
46
+ Amid the swift progress of large language models (LLMs) and their evolution into large multimodal models (LMMs), significant strides have been made in high-resource languages such as English and Chinese. While Arabic LLMs have seen notable progress, Arabic LMMs remain largely unexplored, often narrowly focusing on a few specific aspects of the language and visual understanding. To bridge this gap, we introduce <b><em>AIN - the Arabic Inclusive Multimodal Model-</em></b> designed to excel across diverse domains.
47
+ AIN is an English-Arabic <b>bilingual LMM</b> designed to excel in English and Arabic, leveraging carefully constructed <b>3.6 million</b> high-quality Arabic-English multimodal data samples. AIN demonstrates state-of-the-art Arabic performance, while also possessing strong English-language visual capabilities.
48
+ </p>
49
+ </div>
50
+ </div>
51
+
52
+
53
+
54
+ ## 🌟 Key Features
55
+ - The **first Arabic-centric inclusive Large Multimodal Model (LMM)** trained on **3.6M samples**.
56
+ - Includes **35% authentic Arabic data** within its Arabic data subset.
57
+ - Achieves **superior performance compared to open- and closed-source models** (e.g., GPT-4o) and open-source models (e.g., Qwen2-VL-7B) across tasks such as OCR and specialized domains.
58
+ - Demonstrates **robust bilingual capabilities** (Arabic/English), **validated** through **comprehensive testing** and **human evaluation** across 17 Arab countries.
59
+ - Exhibits **advanced cultural understanding** and domain expertise in fields such as **medical imaging**, **agriculture**, and **scientific visualization**.
60
+
61
+
62
+ <p align="center">
63
+ <img src="assets_hf/intro_bar.png" width="70%" alt="intro_bar" style="margin-right: 2px";/>
64
+ <h6>
65
+ <em> <b>Figure 1.</b> Comparative performance of AIN-7B against other models across key domains, including OCR & Document Understanding, Remote Sensing, Agricultural Understanding, and overall performance across all domains. </em>
66
+ </h6>
67
+ </p>
68
+
69
+ <p align="center" >
70
+ <img src="assets_hf/radar_chart.png" width="52%" alt="radar_chart" style="margin-right: 2px";/>
71
+ <h6>
72
+ <em> <b>Figure 2.</b> showcases a comprehensive performance analysis of AIN-7B across CAMEL-Bench domains, comparing it with prominent closed-source models as well as open-source counterparts. <strong>OCR:</strong> "OCR & Document Understanding", <strong>Video:</strong> "General Video & Multi-Image Understanding", <strong>RS:</strong> "Remote Sensing Understanding", <strong>CDT:</strong> "Chart, Diagram & Table Understanding", <strong>Agro.:</strong> "Agricultural Image Understanding", <strong>Cultural:</strong> "Cultural-Specific Understanding", <strong>Medical:</strong> "Medical Image Understanding".
73
+ </em>
74
+ </h6>
75
+
76
+ ---
77
+ ## ⚖️ Quick Start
78
+ Please install the qwen vision kit. This includes base64, URLs, and interleaved images and videos. You can install it using the following command:
79
+
80
+ ```bash
81
+ pip install qwen-vl-utils
82
+ ```
83
+
84
+ Here we show a code snippet to show you how to use the chat model with `transformers` and `qwen_vl_utils`:
85
+
86
+ ```python
87
+ from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
88
+ from qwen_vl_utils import process_vision_info
89
+ # default: Load the model on the available device(s)
90
+ model = Qwen2VLForConditionalGeneration.from_pretrained(
91
+ "MBZUAI/AIN", torch_dtype="auto", device_map="auto"
92
+ )
93
+ # We recommend enabling flash_attention_2 for better acceleration and memory saving, especially in multi-image and video scenarios.
94
+ # model = Qwen2VLForConditionalGeneration.from_pretrained(
95
+ # "MBZUAI/AIN",
96
+ # torch_dtype=torch.bfloat16,
97
+ # attn_implementation="flash_attention_2",
98
+ # device_map="auto",
99
+ # )
100
+ # default processer
101
+ processor = AutoProcessor.from_pretrained("MBZUAI/AIN")
102
+ # The default range for the number of visual tokens per image in the model is 4-16384. You can set min_pixels and max_pixels according to your needs, such as a token count range of 256-1280, to balance speed and memory usage.
103
+ # min_pixels = 256*28*28
104
+ # max_pixels = 1280*28*28
105
+ # processor = AutoProcessor.from_pretrained("MBZUAI/AIN", min_pixels=min_pixels, max_pixels=max_pixels)
106
+ messages = [
107
+ {
108
+ "role": "user",
109
+ "content": [
110
+ {
111
+ "type": "image",
112
+ "image": "https://huggingface.co/MBZUAI/AIN/resolve/main/assets_hf/demo_image.jpeg",
113
+ },
114
+ {"type": "text", "text": "يرجى وصف هذه الصورة."},
115
+ ],
116
+ }
117
+ ]
118
+ # Preparation for inference
119
+ text = processor.apply_chat_template(
120
+ messages, tokenize=False, add_generation_prompt=True
121
+ )
122
+ image_inputs, video_inputs = process_vision_info(messages)
123
+ inputs = processor(
124
+ text=[text],
125
+ images=image_inputs,
126
+ videos=video_inputs,
127
+ padding=True,
128
+ return_tensors="pt",
129
+ )
130
+ inputs = inputs.to("cuda")
131
+ # Inference: Generation of the output
132
+ generated_ids = model.generate(**inputs, max_new_tokens=128)
133
+ generated_ids_trimmed = [
134
+ out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
135
+ ]
136
+ output_text = processor.batch_decode(
137
+ generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
138
+ )
139
+ print(output_text)
140
+ ```
141
+ <details>
142
+ <summary>Without qwen_vl_utils</summary>
143
+
144
+ ```python
145
+ from PIL import Image
146
+ import requests
147
+ import torch
148
+ from torchvision import io
149
+ from typing import Dict
150
+ from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
151
+ # Load the model in half-precision on the available device(s)
152
+ model = Qwen2VLForConditionalGeneration.from_pretrained(
153
+ "MBZUAI/AIN", torch_dtype="auto", device_map="auto"
154
+ )
155
+ processor = AutoProcessor.from_pretrained("MBZUAI/AIN")
156
+ # Image
157
+ url = "https://huggingface.co/MBZUAI/AIN/resolve/main/assets_hf/demo_image.jpeg"
158
+ image = Image.open(requests.get(url, stream=True).raw)
159
+ conversation = [
160
+ {
161
+ "role": "user",
162
+ "content": [
163
+ {
164
+ "type": "image",
165
+ },
166
+ {"type": "text", "text": "Describe this image in Arabic."},
167
+ ],
168
+ }
169
+ ]
170
+ # Preprocess the inputs
171
+ text_prompt = processor.apply_chat_template(conversation, add_generation_prompt=True)
172
+ # Excepted output: '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>Describe this image.<|im_end|>\n<|im_start|>assistant\n'
173
+ inputs = processor(
174
+ text=[text_prompt], images=[image], padding=True, return_tensors="pt"
175
+ )
176
+ inputs = inputs.to("cuda")
177
+ # Inference: Generation of the output
178
+ output_ids = model.generate(**inputs, max_new_tokens=128)
179
+ generated_ids = [
180
+ output_ids[len(input_ids) :]
181
+ for input_ids, output_ids in zip(inputs.input_ids, output_ids)
182
+ ]
183
+ output_text = processor.batch_decode(
184
+ generated_ids, skip_special_tokens=True, clean_up_tokenization_spaces=True
185
+ )
186
+ print(output_text)
187
+ ```
188
+ </details>
189
+ <details>
190
+ <summary>Multi image inference</summary>
191
+
192
+ ```python
193
+ # Messages containing multiple images and a text query
194
+ messages = [
195
+ {
196
+ "role": "user",
197
+ "content": [
198
+ {"type": "image", "image": "file:///path/to/image1.jpg"},
199
+ {"type": "image", "image": "file:///path/to/image2.jpg"},
200
+ {"type": "text", "text": "Identify the similarities between these images."},
201
+ ],
202
+ }
203
+ ]
204
+ # Preparation for inference
205
+ text = processor.apply_chat_template(
206
+ messages, tokenize=False, add_generation_prompt=True
207
+ )
208
+ image_inputs, video_inputs = process_vision_info(messages)
209
+ inputs = processor(
210
+ text=[text],
211
+ images=image_inputs,
212
+ videos=video_inputs,
213
+ padding=True,
214
+ return_tensors="pt",
215
+ )
216
+ inputs = inputs.to("cuda")
217
+ # Inference
218
+ generated_ids = model.generate(**inputs, max_new_tokens=128)
219
+ generated_ids_trimmed = [
220
+ out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
221
+ ]
222
+ output_text = processor.batch_decode(
223
+ generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
224
+ )
225
+ print(output_text)
226
+ ```
227
+ </details>
228
+
229
+ <details>
230
+ <summary>Video inference</summary>
231
+
232
+ ```python
233
+ # Messages containing a images list as a video and a text query
234
+ messages = [
235
+ {
236
+ "role": "user",
237
+ "content": [
238
+ {
239
+ "type": "video",
240
+ "video": [
241
+ "file:///path/to/frame1.jpg",
242
+ "file:///path/to/frame2.jpg",
243
+ "file:///path/to/frame3.jpg",
244
+ "file:///path/to/frame4.jpg",
245
+ ],
246
+ "fps": 1.0,
247
+ },
248
+ {"type": "text", "text": "Describe this video."},
249
+ ],
250
+ }
251
+ ]
252
+ # Messages containing a video and a text query
253
+ messages = [
254
+ {
255
+ "role": "user",
256
+ "content": [
257
+ {
258
+ "type": "video",
259
+ "video": "file:///path/to/video1.mp4",
260
+ "max_pixels": 360 * 420,
261
+ "fps": 1.0,
262
+ },
263
+ {"type": "text", "text": "Describe this video."},
264
+ ],
265
+ }
266
+ ]
267
+ # Preparation for inference
268
+ text = processor.apply_chat_template(
269
+ messages, tokenize=False, add_generation_prompt=True
270
+ )
271
+ image_inputs, video_inputs = process_vision_info(messages)
272
+ inputs = processor(
273
+ text=[text],
274
+ images=image_inputs,
275
+ videos=video_inputs,
276
+ padding=True,
277
+ return_tensors="pt",
278
+ )
279
+ inputs = inputs.to("cuda")
280
+ # Inference
281
+ generated_ids = model.generate(**inputs, max_new_tokens=128)
282
+ generated_ids_trimmed = [
283
+ out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
284
+ ]
285
+ output_text = processor.batch_decode(
286
+ generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
287
+ )
288
+ print(output_text)
289
+ ```
290
+ </details>
291
+
292
+ <details>
293
+ <summary>Batch inference</summary>
294
+
295
+ ```python
296
+ # Sample messages for batch inference
297
+ messages1 = [
298
+ {
299
+ "role": "user",
300
+ "content": [
301
+ {"type": "image", "image": "file:///path/to/image1.jpg"},
302
+ {"type": "image", "image": "file:///path/to/image2.jpg"},
303
+ {"type": "text", "text": "What are the common elements in these pictures?"},
304
+ ],
305
+ }
306
+ ]
307
+ messages2 = [
308
+ {"role": "system", "content": "You are a helpful assistant."},
309
+ {"role": "user", "content": "Who are you?"},
310
+ ]
311
+ # Combine messages for batch processing
312
+ messages = [messages1, messages1]
313
+ # Preparation for batch inference
314
+ texts = [
315
+ processor.apply_chat_template(msg, tokenize=False, add_generation_prompt=True)
316
+ for msg in messages
317
+ ]
318
+ image_inputs, video_inputs = process_vision_info(messages)
319
+ inputs = processor(
320
+ text=texts,
321
+ images=image_inputs,
322
+ videos=video_inputs,
323
+ padding=True,
324
+ return_tensors="pt",
325
+ )
326
+ inputs = inputs.to("cuda")
327
+ # Batch Inference
328
+ generated_ids = model.generate(**inputs, max_new_tokens=128)
329
+ generated_ids_trimmed = [
330
+ out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
331
+ ]
332
+ output_texts = processor.batch_decode(
333
+ generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
334
+ )
335
+ print(output_texts)
336
+ ```
337
+ </details>
338
+
339
+ ### More Usage Tips
340
+
341
+ For input images, we support local files, base64, and URLs. For videos, we currently only support local files.
342
+
343
+ ```python
344
+ # You can directly insert a local file path, a URL, or a base64-encoded image into the position where you want in the text.
345
+ ## Local file path
346
+ messages = [
347
+ {
348
+ "role": "user",
349
+ "content": [
350
+ {"type": "image", "image": "file:///path/to/your/image.jpg"},
351
+ {"type": "text", "text": "Describe this image."},
352
+ ],
353
+ }
354
+ ]
355
+ ## Image URL
356
+ messages = [
357
+ {
358
+ "role": "user",
359
+ "content": [
360
+ {"type": "image", "image": "http://path/to/your/image.jpg"},
361
+ {"type": "text", "text": "Describe this image."},
362
+ ],
363
+ }
364
+ ]
365
+ ## Base64 encoded image
366
+ messages = [
367
+ {
368
+ "role": "user",
369
+ "content": [
370
+ {"type": "image", "image": "data:image;base64,/9j/..."},
371
+ {"type": "text", "text": "Describe this image."},
372
+ ],
373
+ }
374
+ ]
375
+ ```
376
+ #### Image Resolution for performance boost
377
+
378
+ The model supports a wide range of resolution inputs. By default, it uses the native resolution for input, but higher resolutions can enhance performance at the cost of more computation. Users can set the minimum and maximum number of pixels to achieve an optimal configuration for their needs, such as a token count range of 256-1280, to balance speed and memory usage.
379
+
380
+ ```python
381
+ min_pixels = 256 * 28 * 28
382
+ max_pixels = 1280 * 28 * 28
383
+ processor = AutoProcessor.from_pretrained(
384
+ "MBZUAI/AIN", min_pixels=min_pixels, max_pixels=max_pixels
385
+ )
386
+ ```
387
+
388
+ Besides, We provide two methods for fine-grained control over the image size input to the model:
389
+
390
+ 1. Define min_pixels and max_pixels: Images will be resized to maintain their aspect ratio within the range of min_pixels and max_pixels.
391
+
392
+ 2. Specify exact dimensions: Directly set `resized_height` and `resized_width`. These values will be rounded to the nearest multiple of 28.
393
+
394
+ ```python
395
+ # min_pixels and max_pixels
396
+ messages = [
397
+ {
398
+ "role": "user",
399
+ "content": [
400
+ {
401
+ "type": "image",
402
+ "image": "file:///path/to/your/image.jpg",
403
+ "resized_height": 280,
404
+ "resized_width": 420,
405
+ },
406
+ {"type": "text", "text": "Describe this image."},
407
+ ],
408
+ }
409
+ ]
410
+ # resized_height and resized_width
411
+ messages = [
412
+ {
413
+ "role": "user",
414
+ "content": [
415
+ {
416
+ "type": "image",
417
+ "image": "file:///path/to/your/image.jpg",
418
+ "min_pixels": 50176,
419
+ "max_pixels": 50176,
420
+ },
421
+ {"type": "text", "text": "Describe this image."},
422
+ ],
423
+ }
424
+ ]
425
+ ```
426
+
427
+ ---
428
+ ## ⚖️ Quantitative Evaluation and Results
429
+ AIN demonstrates state-of-the-art performance across diverse domains, surpassing both open- and closed-source models. Notably, it achieves an aggregate performance score of 63.77%, with significant gains in OCR, remote sensing, and agricultural image understanding.
430
+
431
+ <div align="center" >
432
+ <table>
433
+ <caption>
434
+ <h6>
435
+ <strong>Table 1. Performance comparison of AIN and different closed- and open-source LMMs across CAMEL-Bench domains.</strong>
436
+ <br> <em>Best performance is marked with 🥇; second-best is 🥈.</em>
437
+ <strong>OCR</strong>: "OCR & Document Understanding",
438
+ <strong>Video</strong>: "General Video & Multi-Image Understanding",
439
+ <strong>RS</strong>: "Remote Sensing Understanding",
440
+ <strong>CDT</strong>: "Chart, Diagram & Table Understanding",
441
+ <strong>Agro.</strong>: "Agricultural Image Understanding",
442
+ <strong>Cult.</strong>: "Cultural-Specific Understanding",
443
+ <strong>Med.</strong>: "Medical Image Understanding".
444
+ </h6>
445
+ </caption>
446
+ <thead>
447
+ <tr style="background-color: #e0e0e0;">
448
+ <th>Models</th>
449
+ <th>VQA</th>
450
+ <th>OCR</th>
451
+ <th>Video</th>
452
+ <th>RS</th>
453
+ <th>CDT</th>
454
+ <th>Agro.</th>
455
+ <th>Cult.</th>
456
+ <th>Med.</th>
457
+ <th style="background-color: #d0d0d0;">Total</th>
458
+ </tr>
459
+ </thead>
460
+ <tbody>
461
+ <tr>
462
+ <td>GPT-4o</td>
463
+ <td>🥈55.15</td>
464
+ <td>🥈54.98</td>
465
+ <td>🥇69.65</td>
466
+ <td>🥈27.36</td>
467
+ <td>🥈62.35</td>
468
+ <td>🥈80.75</td>
469
+ <td>🥇80.86</td>
470
+ <td>🥇49.91</td>
471
+ <td style="background-color: #d0d0d0;">🥈60.13</td>
472
+ </tr>
473
+ <tr>
474
+ <td>GPT-4o-mini</td>
475
+ <td>48.83</td>
476
+ <td>39.38</td>
477
+ <td>🥈66.28</td>
478
+ <td>16.93</td>
479
+ <td>56.37</td>
480
+ <td>78.80</td>
481
+ <td>65.92</td>
482
+ <td>🥈47.37</td>
483
+ <td style="background-color: #d0d0d0;">52.49</td>
484
+ </tr>
485
+ <tr>
486
+ <td>Gemini-1.5-Pro</td>
487
+ <td>46.68</td>
488
+ <td>28.68</td>
489
+ <td>42.95</td>
490
+ <td>17.07</td>
491
+ <td>47.06</td>
492
+ <td>72.14</td>
493
+ <td>56.24</td>
494
+ <td>33.78</td>
495
+ <td style="background-color: #d0d0d0;">52.38</td>
496
+ </tr>
497
+ <tr>
498
+ <td>Gemini-1.5-flash</td>
499
+ <td>45.59</td>
500
+ <td>27.58</td>
501
+ <td>53.31</td>
502
+ <td>14.95</td>
503
+ <td>48.26</td>
504
+ <td>76.07</td>
505
+ <td>46.54</td>
506
+ <td>42.87</td>
507
+ <td style="background-color: #d0d0d0;">44.40</td>
508
+ </tr>
509
+ <tr>
510
+ <td>InternVL-8B </td>
511
+ <td>30.41 </td>
512
+ <td>15.91 </td>
513
+ <td>51.42 </td>
514
+ <td>5.36 </td>
515
+ <td>30.27 </td>
516
+ <td>44.47 </td>
517
+ <td>20.88 </td>
518
+ <td>29.48 </td>
519
+ <td style="background-color: #d0d0d0;">28.52 </td>
520
+ </tr>
521
+ <tr>
522
+ <td>InternVL2.5-1B </td>
523
+ <td>27.22 </td>
524
+ <td>19.45 </td>
525
+ <td>38.20 </td>
526
+ <td>3.39 </td>
527
+ <td>30.75 </td>
528
+ <td>39.53 </td>
529
+ <td>35.68 </td>
530
+ <td>21.27 </td>
531
+ <td style="background-color: #d0d0d0;">26.94 </td>
532
+ </tr>
533
+ <tr>
534
+ <td>Qwen-VL-2B </td>
535
+ <td>41.02 </td>
536
+ <td>22.93 </td>
537
+ <td>38.90 </td>
538
+ <td>12.56 </td>
539
+ <td>27.83 </td>
540
+ <td>52.02 </td>
541
+ <td>34.28 </td>
542
+ <td>29.12 </td>
543
+ <td style="background-color: #d0d0d0;">32.33 </td>
544
+ </tr>
545
+ <tr>
546
+ <td>Qwen2-VL-7B </td>
547
+ <td>48.76 </td>
548
+ <td>42.73 </td>
549
+ <td>61.97 </td>
550
+ <td>21.30 </td>
551
+ <td>54.67 </td>
552
+ <td>79.32 </td>
553
+ <td>75.96 </td>
554
+ <td>35.81 </td>
555
+ <td style="background-color: #d0d0d0;">52.57 </td>
556
+ </tr>
557
+ <tr>
558
+ <td>AIN-7B <em>(ours)</em> </td>
559
+ <td>🥇56.78 </td>
560
+ <td>🥇72.35 </td>
561
+ <td>64.09 </td>
562
+ <td>🥇45.92 </td>
563
+ <td>🥇64.10 </td>
564
+ <td>🥇85.05 </td>
565
+ <td>🥈78.09 </td>
566
+ <td>43.77 </td>
567
+ <td style="background-color: #d0d0d0;">🏆63.77 </td>
568
+ </tr>
569
+ </tbody>
570
+ </table>
571
+ </div>
572
+
573
+ ---
574
+ ## 🎯 Qualitative Evaluation
575
+ The qualitative evaluation showcases AIN's advanced capabilities in handling diverse, complex tasks, including OCR, medical imaging, remote sensing, and cultural-specific understanding, with remarkable precision and contextual relevance. Unlike GPT-4o and LLaVA, AIN demonstrates superior performance in identifying intricate details and maintaining accuracy across varied query formats and multi-domain challenges.
576
+
577
+ <div align="center">
578
+ <img src="assets_hf/qualitative.png" width="75%" alt="qualitative" />
579
+ <h6>
580
+ <em> <b>Figure 3.</b> Qualitative examples showcasing AIN-7B’s capabilities across various domains, including general VQA, OCR & Document Understanding, Remote Sensing, Medical Imaging, Agricultural Understanding, and Cultural-Specific tasks. </em>
581
+ </h6>
582
+ </div>
583
+
584
+ ---
585
+ ## 🧐 Data Verification and Toxicity Filtering
586
+ A multi-step verification pipeline was implemented to ensure high-quality translations and safe visual data. Translation accuracy was assessed through human evaluation, where native Arabic speakers rated outputs against reference translations, and semantic similarity checks were conducted using **LaBSE**. Additionally, translated samples were reverse-translated and validated using **BLEU, METEOR, and ROUGE scores** to measure correctness, correlation, and overlap. For visual data, toxicity filtering was applied using **LLavaGuard’s safety policies and GPT-4o**, identifying and removing unsafe content related to violence, substance abuse, and harmful imagery, ensuring compliance with ethical AI standards.
587
+
588
+ <p align="center">
589
+ <img src="assets_hf/verify_pipeline.png" width="75%" alt="verify" style="margin-right: 2px";/>
590
+ <h6>
591
+ <em> <b>Figure 4.</b> Data verification and filtering pipeline for textual and visual data, ensuring high-quality training data through semantic similarity checks, translation quality evaluations, and toxicity screening for safety compliance. </em>
592
+ </h6>
593
+ </p>
594
+ <p align="center">
595
+ <img src="assets_hf/toxicity.png" width=48%" alt="verify" style="margin-right: 2px";/>
596
+ <h6>
597
+ <em> <b>Figure 5.</b> Distribution of visual data toxicity filtering results, showing that 95% of the data is classified as safe, while 5% is identified as unsafe due to categories like weapons or substance abuse, violence, and animal cruelty. </em>
598
+ </h6>
599
+ </p>
600
+
601
+ ---
602
+
603
+ ## 🔒 License
604
+ This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
605
+
606
+
607
+ ## 💬 Contact us
608
+ For questions or suggestions, feel free to reach out to us on [GitHub Discussions](https://github.com/mbzuai-oryx/AIN/discussions).
609
+
610
+ ---
611
+
612
+ If you use AIN in your research, please cite our work as follows:
613
+
614
+ ```
615
+ @misc{heakl2025ainarabicinclusivelarge,
616
+ title={AIN: The Arabic INclusive Large Multimodal Model},
617
+ author={Ahmed Heakl and Sara Ghaboura and Omkar Thawkar and Fahad Shahbaz Khan and Hisham Cholakkal and Rao Muhammad Anwer and Salman Khan},
618
+ year={2025},
619
+ eprint={2502.00094},
620
+ url={https://arxiv.org/abs/2502.00094},
621
+ ```
622
+ ---
added_tokens.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ }
assets_hf/.DS_Store ADDED
Binary file (6.15 kB). View file
 
assets_hf/AIN.png ADDED

Git LFS Details

  • SHA256: 2688a836c3fd0abc0c3cf734ed9550592c9cb276dfc80b3185b66c00bead907c
  • Pointer size: 131 Bytes
  • Size of remote file: 758 kB
assets_hf/Eval_CAMEL.png ADDED

Git LFS Details

  • SHA256: 61562a4a14c1efec1967f127f793746e82f5e997de27af5b789a3e717219b182
  • Pointer size: 131 Bytes
  • Size of remote file: 159 kB
assets_hf/ain_can_see.png ADDED

Git LFS Details

  • SHA256: ee835a86e0f920155aab4d8a6f33b2eb370611157a3dd18770f7b90783d87986
  • Pointer size: 131 Bytes
  • Size of remote file: 505 kB
assets_hf/demo_image.jpeg ADDED

Git LFS Details

  • SHA256: 9eeaa87013b4e800930e8a411b58ff9e2fd5383906b1a022f4a712720af34cc2
  • Pointer size: 131 Bytes
  • Size of remote file: 496 kB
assets_hf/intro_bar.png ADDED
assets_hf/qualitative.png ADDED

Git LFS Details

  • SHA256: 4ffc3aad5b0269174318094029e4c2222baffb925544070b15242578bb4edc2c
  • Pointer size: 131 Bytes
  • Size of remote file: 487 kB
assets_hf/radar_chart.png ADDED

Git LFS Details

  • SHA256: 030ee23b6297b70c3cc729b74d62937fb796d910ad32883bfe0e80c9f0497dc8
  • Pointer size: 131 Bytes
  • Size of remote file: 220 kB
assets_hf/toxicity.png ADDED

Git LFS Details

  • SHA256: 465b28e4f4a87c27eb74fcb6ba88326b75b0d79214fe73ef443f0f01e205826d
  • Pointer size: 131 Bytes
  • Size of remote file: 414 kB
assets_hf/verify_pipeline.png ADDED

Git LFS Details

  • SHA256: ed49180a5e33b6a27965611495b3cdaeee47b0570167aa0788b2c09c66106540
  • Pointer size: 131 Bytes
  • Size of remote file: 156 kB
chat_template.json ADDED
@@ -0,0 +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
+ }
config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Qwen/Qwen2-VL-7B-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": 3584,
11
+ "image_token_id": 151655,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 18944,
14
+ "max_position_embeddings": 32768,
15
+ "max_window_layers": 28,
16
+ "model_type": "qwen2_vl",
17
+ "num_attention_heads": 28,
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
+ "tie_word_embeddings": false,
33
+ "torch_dtype": "bfloat16",
34
+ "transformers_version": "4.46.1",
35
+ "use_cache": false,
36
+ "use_sliding_window": false,
37
+ "video_token_id": 151656,
38
+ "vision_config": {
39
+ "in_chans": 3,
40
+ "model_type": "qwen2_vl",
41
+ "spatial_patch_size": 14
42
+ },
43
+ "vision_end_token_id": 151653,
44
+ "vision_start_token_id": 151652,
45
+ "vision_token_id": 151654,
46
+ "vocab_size": 152064
47
+ }
generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "temperature": 0.01,
10
+ "top_k": 1,
11
+ "top_p": 0.001,
12
+ "transformers_version": "4.46.1"
13
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model-00001-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:521a7f855fe7007de55d5ee50db1454b0f25e5e4c0315045337c9fefe39e0816
3
+ size 4949146168
model-00002-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:038ebf8d995aa6ceb842f1980340b96ec1ecbda1ad3f23a7ffb1ae164f57dd73
3
+ size 4984124272
model-00003-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce491efce017d261faef2c0f80aa247bbc19a9ee3f1c235f08fc770604cdc474
3
+ size 4932743936
model-00004-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b6f52b6deda6d0d3bada4f5b3dbf5dab045ce67a82d9e72e018ac99950883cc
3
+ size 4998852296
model-00005-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92e24b3cf9fb877bbf1e89fe109996d72509e2ae4235d0c8ca204925f4ee1f1b
3
+ size 4984124336
model-00006-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3828b394021b0632e84c638e081ec3f349bde0936c87db1e9045b066eb63ccfc
3
+ size 4932743992
model-00007-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d0311852fc0d983b2608a62d455992db52342d6c488d48ddc4f8a8aac820fcb
3
+ size 3383846800
model.safetensors.index.json ADDED
@@ -0,0 +1,737 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 33165502464
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "model-00007-of-00007.safetensors",
7
+ "model.embed_tokens.weight": "model-00001-of-00007.safetensors",
8
+ "model.layers.0.input_layernorm.weight": "model-00002-of-00007.safetensors",
9
+ "model.layers.0.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
10
+ "model.layers.0.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
11
+ "model.layers.0.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
12
+ "model.layers.0.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
13
+ "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
14
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
15
+ "model.layers.0.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
16
+ "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
17
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
18
+ "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
19
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
20
+ "model.layers.1.input_layernorm.weight": "model-00002-of-00007.safetensors",
21
+ "model.layers.1.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
22
+ "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
23
+ "model.layers.1.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
24
+ "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
25
+ "model.layers.1.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
26
+ "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
27
+ "model.layers.1.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
28
+ "model.layers.1.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
29
+ "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
30
+ "model.layers.1.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
31
+ "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
32
+ "model.layers.10.input_layernorm.weight": "model-00004-of-00007.safetensors",
33
+ "model.layers.10.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
34
+ "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
35
+ "model.layers.10.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
36
+ "model.layers.10.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
37
+ "model.layers.10.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
38
+ "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
39
+ "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
40
+ "model.layers.10.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
41
+ "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
42
+ "model.layers.10.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
43
+ "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
44
+ "model.layers.11.input_layernorm.weight": "model-00004-of-00007.safetensors",
45
+ "model.layers.11.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
46
+ "model.layers.11.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
47
+ "model.layers.11.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
48
+ "model.layers.11.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
49
+ "model.layers.11.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
50
+ "model.layers.11.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
51
+ "model.layers.11.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
52
+ "model.layers.11.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
53
+ "model.layers.11.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
54
+ "model.layers.11.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
55
+ "model.layers.11.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
56
+ "model.layers.12.input_layernorm.weight": "model-00004-of-00007.safetensors",
57
+ "model.layers.12.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
58
+ "model.layers.12.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
59
+ "model.layers.12.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
60
+ "model.layers.12.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
61
+ "model.layers.12.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
62
+ "model.layers.12.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
63
+ "model.layers.12.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
64
+ "model.layers.12.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
65
+ "model.layers.12.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
66
+ "model.layers.12.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
67
+ "model.layers.12.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
68
+ "model.layers.13.input_layernorm.weight": "model-00004-of-00007.safetensors",
69
+ "model.layers.13.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
70
+ "model.layers.13.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
71
+ "model.layers.13.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
72
+ "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
73
+ "model.layers.13.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
74
+ "model.layers.13.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
75
+ "model.layers.13.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
76
+ "model.layers.13.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
77
+ "model.layers.13.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
78
+ "model.layers.13.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
79
+ "model.layers.13.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
80
+ "model.layers.14.input_layernorm.weight": "model-00004-of-00007.safetensors",
81
+ "model.layers.14.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
82
+ "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
83
+ "model.layers.14.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
84
+ "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
85
+ "model.layers.14.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
86
+ "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
87
+ "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
88
+ "model.layers.14.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
89
+ "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
90
+ "model.layers.14.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
91
+ "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
92
+ "model.layers.15.input_layernorm.weight": "model-00004-of-00007.safetensors",
93
+ "model.layers.15.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
94
+ "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
95
+ "model.layers.15.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
96
+ "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
97
+ "model.layers.15.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
98
+ "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
99
+ "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
100
+ "model.layers.15.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
101
+ "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
102
+ "model.layers.15.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
103
+ "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
104
+ "model.layers.16.input_layernorm.weight": "model-00005-of-00007.safetensors",
105
+ "model.layers.16.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
106
+ "model.layers.16.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
107
+ "model.layers.16.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
108
+ "model.layers.16.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
109
+ "model.layers.16.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
110
+ "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
111
+ "model.layers.16.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
112
+ "model.layers.16.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
113
+ "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
114
+ "model.layers.16.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
115
+ "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
116
+ "model.layers.17.input_layernorm.weight": "model-00005-of-00007.safetensors",
117
+ "model.layers.17.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
118
+ "model.layers.17.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
119
+ "model.layers.17.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
120
+ "model.layers.17.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
121
+ "model.layers.17.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
122
+ "model.layers.17.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
123
+ "model.layers.17.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
124
+ "model.layers.17.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
125
+ "model.layers.17.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
126
+ "model.layers.17.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
127
+ "model.layers.17.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
128
+ "model.layers.18.input_layernorm.weight": "model-00005-of-00007.safetensors",
129
+ "model.layers.18.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
130
+ "model.layers.18.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
131
+ "model.layers.18.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
132
+ "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
133
+ "model.layers.18.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
134
+ "model.layers.18.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
135
+ "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
136
+ "model.layers.18.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
137
+ "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
138
+ "model.layers.18.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
139
+ "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
140
+ "model.layers.19.input_layernorm.weight": "model-00005-of-00007.safetensors",
141
+ "model.layers.19.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
142
+ "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
143
+ "model.layers.19.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
144
+ "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
145
+ "model.layers.19.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
146
+ "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
147
+ "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
148
+ "model.layers.19.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
149
+ "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
150
+ "model.layers.19.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
151
+ "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
152
+ "model.layers.2.input_layernorm.weight": "model-00002-of-00007.safetensors",
153
+ "model.layers.2.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
154
+ "model.layers.2.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
155
+ "model.layers.2.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
156
+ "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
157
+ "model.layers.2.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
158
+ "model.layers.2.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
159
+ "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
160
+ "model.layers.2.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
161
+ "model.layers.2.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
162
+ "model.layers.2.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
163
+ "model.layers.2.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
164
+ "model.layers.20.input_layernorm.weight": "model-00005-of-00007.safetensors",
165
+ "model.layers.20.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
166
+ "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
167
+ "model.layers.20.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
168
+ "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
169
+ "model.layers.20.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
170
+ "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
171
+ "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
172
+ "model.layers.20.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
173
+ "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
174
+ "model.layers.20.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
175
+ "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
176
+ "model.layers.21.input_layernorm.weight": "model-00006-of-00007.safetensors",
177
+ "model.layers.21.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
178
+ "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
179
+ "model.layers.21.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
180
+ "model.layers.21.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
181
+ "model.layers.21.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
182
+ "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
183
+ "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
184
+ "model.layers.21.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
185
+ "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
186
+ "model.layers.21.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
187
+ "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
188
+ "model.layers.22.input_layernorm.weight": "model-00006-of-00007.safetensors",
189
+ "model.layers.22.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
190
+ "model.layers.22.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
191
+ "model.layers.22.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
192
+ "model.layers.22.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
193
+ "model.layers.22.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
194
+ "model.layers.22.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
195
+ "model.layers.22.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
196
+ "model.layers.22.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
197
+ "model.layers.22.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
198
+ "model.layers.22.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
199
+ "model.layers.22.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
200
+ "model.layers.23.input_layernorm.weight": "model-00006-of-00007.safetensors",
201
+ "model.layers.23.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
202
+ "model.layers.23.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
203
+ "model.layers.23.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
204
+ "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
205
+ "model.layers.23.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
206
+ "model.layers.23.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
207
+ "model.layers.23.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
208
+ "model.layers.23.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
209
+ "model.layers.23.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
210
+ "model.layers.23.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
211
+ "model.layers.23.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
212
+ "model.layers.24.input_layernorm.weight": "model-00006-of-00007.safetensors",
213
+ "model.layers.24.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
214
+ "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
215
+ "model.layers.24.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
216
+ "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
217
+ "model.layers.24.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
218
+ "model.layers.24.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
219
+ "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
220
+ "model.layers.24.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
221
+ "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
222
+ "model.layers.24.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
223
+ "model.layers.24.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
224
+ "model.layers.25.input_layernorm.weight": "model-00006-of-00007.safetensors",
225
+ "model.layers.25.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
226
+ "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
227
+ "model.layers.25.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
228
+ "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
229
+ "model.layers.25.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
230
+ "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
231
+ "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
232
+ "model.layers.25.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
233
+ "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
234
+ "model.layers.25.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
235
+ "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
236
+ "model.layers.26.input_layernorm.weight": "model-00007-of-00007.safetensors",
237
+ "model.layers.26.mlp.down_proj.weight": "model-00007-of-00007.safetensors",
238
+ "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
239
+ "model.layers.26.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
240
+ "model.layers.26.post_attention_layernorm.weight": "model-00007-of-00007.safetensors",
241
+ "model.layers.26.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
242
+ "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
243
+ "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
244
+ "model.layers.26.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
245
+ "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
246
+ "model.layers.26.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
247
+ "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
248
+ "model.layers.27.input_layernorm.weight": "model-00007-of-00007.safetensors",
249
+ "model.layers.27.mlp.down_proj.weight": "model-00007-of-00007.safetensors",
250
+ "model.layers.27.mlp.gate_proj.weight": "model-00007-of-00007.safetensors",
251
+ "model.layers.27.mlp.up_proj.weight": "model-00007-of-00007.safetensors",
252
+ "model.layers.27.post_attention_layernorm.weight": "model-00007-of-00007.safetensors",
253
+ "model.layers.27.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
254
+ "model.layers.27.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
255
+ "model.layers.27.self_attn.o_proj.weight": "model-00007-of-00007.safetensors",
256
+ "model.layers.27.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
257
+ "model.layers.27.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
258
+ "model.layers.27.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
259
+ "model.layers.27.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
260
+ "model.layers.3.input_layernorm.weight": "model-00002-of-00007.safetensors",
261
+ "model.layers.3.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
262
+ "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
263
+ "model.layers.3.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
264
+ "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
265
+ "model.layers.3.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
266
+ "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
267
+ "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
268
+ "model.layers.3.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
269
+ "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
270
+ "model.layers.3.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
271
+ "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
272
+ "model.layers.4.input_layernorm.weight": "model-00002-of-00007.safetensors",
273
+ "model.layers.4.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
274
+ "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
275
+ "model.layers.4.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
276
+ "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
277
+ "model.layers.4.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
278
+ "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
279
+ "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
280
+ "model.layers.4.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
281
+ "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
282
+ "model.layers.4.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
283
+ "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
284
+ "model.layers.5.input_layernorm.weight": "model-00003-of-00007.safetensors",
285
+ "model.layers.5.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
286
+ "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
287
+ "model.layers.5.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
288
+ "model.layers.5.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
289
+ "model.layers.5.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
290
+ "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
291
+ "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
292
+ "model.layers.5.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
293
+ "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
294
+ "model.layers.5.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
295
+ "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
296
+ "model.layers.6.input_layernorm.weight": "model-00003-of-00007.safetensors",
297
+ "model.layers.6.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
298
+ "model.layers.6.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
299
+ "model.layers.6.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
300
+ "model.layers.6.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
301
+ "model.layers.6.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
302
+ "model.layers.6.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
303
+ "model.layers.6.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
304
+ "model.layers.6.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
305
+ "model.layers.6.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
306
+ "model.layers.6.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
307
+ "model.layers.6.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
308
+ "model.layers.7.input_layernorm.weight": "model-00003-of-00007.safetensors",
309
+ "model.layers.7.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
310
+ "model.layers.7.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
311
+ "model.layers.7.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
312
+ "model.layers.7.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
313
+ "model.layers.7.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
314
+ "model.layers.7.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
315
+ "model.layers.7.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
316
+ "model.layers.7.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
317
+ "model.layers.7.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
318
+ "model.layers.7.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
319
+ "model.layers.7.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
320
+ "model.layers.8.input_layernorm.weight": "model-00003-of-00007.safetensors",
321
+ "model.layers.8.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
322
+ "model.layers.8.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
323
+ "model.layers.8.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
324
+ "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
325
+ "model.layers.8.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
326
+ "model.layers.8.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
327
+ "model.layers.8.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
328
+ "model.layers.8.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
329
+ "model.layers.8.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
330
+ "model.layers.8.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
331
+ "model.layers.8.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
332
+ "model.layers.9.input_layernorm.weight": "model-00003-of-00007.safetensors",
333
+ "model.layers.9.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
334
+ "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
335
+ "model.layers.9.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
336
+ "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
337
+ "model.layers.9.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
338
+ "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
339
+ "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
340
+ "model.layers.9.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
341
+ "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
342
+ "model.layers.9.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
343
+ "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
344
+ "model.norm.weight": "model-00007-of-00007.safetensors",
345
+ "visual.blocks.0.attn.proj.bias": "model-00001-of-00007.safetensors",
346
+ "visual.blocks.0.attn.proj.weight": "model-00001-of-00007.safetensors",
347
+ "visual.blocks.0.attn.qkv.bias": "model-00001-of-00007.safetensors",
348
+ "visual.blocks.0.attn.qkv.weight": "model-00001-of-00007.safetensors",
349
+ "visual.blocks.0.mlp.fc1.bias": "model-00001-of-00007.safetensors",
350
+ "visual.blocks.0.mlp.fc1.weight": "model-00001-of-00007.safetensors",
351
+ "visual.blocks.0.mlp.fc2.bias": "model-00001-of-00007.safetensors",
352
+ "visual.blocks.0.mlp.fc2.weight": "model-00001-of-00007.safetensors",
353
+ "visual.blocks.0.norm1.bias": "model-00001-of-00007.safetensors",
354
+ "visual.blocks.0.norm1.weight": "model-00001-of-00007.safetensors",
355
+ "visual.blocks.0.norm2.bias": "model-00001-of-00007.safetensors",
356
+ "visual.blocks.0.norm2.weight": "model-00001-of-00007.safetensors",
357
+ "visual.blocks.1.attn.proj.bias": "model-00001-of-00007.safetensors",
358
+ "visual.blocks.1.attn.proj.weight": "model-00001-of-00007.safetensors",
359
+ "visual.blocks.1.attn.qkv.bias": "model-00001-of-00007.safetensors",
360
+ "visual.blocks.1.attn.qkv.weight": "model-00001-of-00007.safetensors",
361
+ "visual.blocks.1.mlp.fc1.bias": "model-00001-of-00007.safetensors",
362
+ "visual.blocks.1.mlp.fc1.weight": "model-00001-of-00007.safetensors",
363
+ "visual.blocks.1.mlp.fc2.bias": "model-00001-of-00007.safetensors",
364
+ "visual.blocks.1.mlp.fc2.weight": "model-00001-of-00007.safetensors",
365
+ "visual.blocks.1.norm1.bias": "model-00001-of-00007.safetensors",
366
+ "visual.blocks.1.norm1.weight": "model-00001-of-00007.safetensors",
367
+ "visual.blocks.1.norm2.bias": "model-00001-of-00007.safetensors",
368
+ "visual.blocks.1.norm2.weight": "model-00001-of-00007.safetensors",
369
+ "visual.blocks.10.attn.proj.bias": "model-00001-of-00007.safetensors",
370
+ "visual.blocks.10.attn.proj.weight": "model-00001-of-00007.safetensors",
371
+ "visual.blocks.10.attn.qkv.bias": "model-00001-of-00007.safetensors",
372
+ "visual.blocks.10.attn.qkv.weight": "model-00001-of-00007.safetensors",
373
+ "visual.blocks.10.mlp.fc1.bias": "model-00001-of-00007.safetensors",
374
+ "visual.blocks.10.mlp.fc1.weight": "model-00001-of-00007.safetensors",
375
+ "visual.blocks.10.mlp.fc2.bias": "model-00001-of-00007.safetensors",
376
+ "visual.blocks.10.mlp.fc2.weight": "model-00001-of-00007.safetensors",
377
+ "visual.blocks.10.norm1.bias": "model-00001-of-00007.safetensors",
378
+ "visual.blocks.10.norm1.weight": "model-00001-of-00007.safetensors",
379
+ "visual.blocks.10.norm2.bias": "model-00001-of-00007.safetensors",
380
+ "visual.blocks.10.norm2.weight": "model-00001-of-00007.safetensors",
381
+ "visual.blocks.11.attn.proj.bias": "model-00001-of-00007.safetensors",
382
+ "visual.blocks.11.attn.proj.weight": "model-00001-of-00007.safetensors",
383
+ "visual.blocks.11.attn.qkv.bias": "model-00001-of-00007.safetensors",
384
+ "visual.blocks.11.attn.qkv.weight": "model-00001-of-00007.safetensors",
385
+ "visual.blocks.11.mlp.fc1.bias": "model-00001-of-00007.safetensors",
386
+ "visual.blocks.11.mlp.fc1.weight": "model-00001-of-00007.safetensors",
387
+ "visual.blocks.11.mlp.fc2.bias": "model-00001-of-00007.safetensors",
388
+ "visual.blocks.11.mlp.fc2.weight": "model-00001-of-00007.safetensors",
389
+ "visual.blocks.11.norm1.bias": "model-00001-of-00007.safetensors",
390
+ "visual.blocks.11.norm1.weight": "model-00001-of-00007.safetensors",
391
+ "visual.blocks.11.norm2.bias": "model-00001-of-00007.safetensors",
392
+ "visual.blocks.11.norm2.weight": "model-00001-of-00007.safetensors",
393
+ "visual.blocks.12.attn.proj.bias": "model-00001-of-00007.safetensors",
394
+ "visual.blocks.12.attn.proj.weight": "model-00001-of-00007.safetensors",
395
+ "visual.blocks.12.attn.qkv.bias": "model-00001-of-00007.safetensors",
396
+ "visual.blocks.12.attn.qkv.weight": "model-00001-of-00007.safetensors",
397
+ "visual.blocks.12.mlp.fc1.bias": "model-00001-of-00007.safetensors",
398
+ "visual.blocks.12.mlp.fc1.weight": "model-00001-of-00007.safetensors",
399
+ "visual.blocks.12.mlp.fc2.bias": "model-00001-of-00007.safetensors",
400
+ "visual.blocks.12.mlp.fc2.weight": "model-00001-of-00007.safetensors",
401
+ "visual.blocks.12.norm1.bias": "model-00001-of-00007.safetensors",
402
+ "visual.blocks.12.norm1.weight": "model-00001-of-00007.safetensors",
403
+ "visual.blocks.12.norm2.bias": "model-00001-of-00007.safetensors",
404
+ "visual.blocks.12.norm2.weight": "model-00001-of-00007.safetensors",
405
+ "visual.blocks.13.attn.proj.bias": "model-00001-of-00007.safetensors",
406
+ "visual.blocks.13.attn.proj.weight": "model-00001-of-00007.safetensors",
407
+ "visual.blocks.13.attn.qkv.bias": "model-00001-of-00007.safetensors",
408
+ "visual.blocks.13.attn.qkv.weight": "model-00001-of-00007.safetensors",
409
+ "visual.blocks.13.mlp.fc1.bias": "model-00001-of-00007.safetensors",
410
+ "visual.blocks.13.mlp.fc1.weight": "model-00001-of-00007.safetensors",
411
+ "visual.blocks.13.mlp.fc2.bias": "model-00001-of-00007.safetensors",
412
+ "visual.blocks.13.mlp.fc2.weight": "model-00001-of-00007.safetensors",
413
+ "visual.blocks.13.norm1.bias": "model-00001-of-00007.safetensors",
414
+ "visual.blocks.13.norm1.weight": "model-00001-of-00007.safetensors",
415
+ "visual.blocks.13.norm2.bias": "model-00001-of-00007.safetensors",
416
+ "visual.blocks.13.norm2.weight": "model-00001-of-00007.safetensors",
417
+ "visual.blocks.14.attn.proj.bias": "model-00001-of-00007.safetensors",
418
+ "visual.blocks.14.attn.proj.weight": "model-00001-of-00007.safetensors",
419
+ "visual.blocks.14.attn.qkv.bias": "model-00001-of-00007.safetensors",
420
+ "visual.blocks.14.attn.qkv.weight": "model-00001-of-00007.safetensors",
421
+ "visual.blocks.14.mlp.fc1.bias": "model-00001-of-00007.safetensors",
422
+ "visual.blocks.14.mlp.fc1.weight": "model-00001-of-00007.safetensors",
423
+ "visual.blocks.14.mlp.fc2.bias": "model-00001-of-00007.safetensors",
424
+ "visual.blocks.14.mlp.fc2.weight": "model-00001-of-00007.safetensors",
425
+ "visual.blocks.14.norm1.bias": "model-00001-of-00007.safetensors",
426
+ "visual.blocks.14.norm1.weight": "model-00001-of-00007.safetensors",
427
+ "visual.blocks.14.norm2.bias": "model-00001-of-00007.safetensors",
428
+ "visual.blocks.14.norm2.weight": "model-00001-of-00007.safetensors",
429
+ "visual.blocks.15.attn.proj.bias": "model-00001-of-00007.safetensors",
430
+ "visual.blocks.15.attn.proj.weight": "model-00001-of-00007.safetensors",
431
+ "visual.blocks.15.attn.qkv.bias": "model-00001-of-00007.safetensors",
432
+ "visual.blocks.15.attn.qkv.weight": "model-00001-of-00007.safetensors",
433
+ "visual.blocks.15.mlp.fc1.bias": "model-00001-of-00007.safetensors",
434
+ "visual.blocks.15.mlp.fc1.weight": "model-00001-of-00007.safetensors",
435
+ "visual.blocks.15.mlp.fc2.bias": "model-00001-of-00007.safetensors",
436
+ "visual.blocks.15.mlp.fc2.weight": "model-00001-of-00007.safetensors",
437
+ "visual.blocks.15.norm1.bias": "model-00001-of-00007.safetensors",
438
+ "visual.blocks.15.norm1.weight": "model-00001-of-00007.safetensors",
439
+ "visual.blocks.15.norm2.bias": "model-00001-of-00007.safetensors",
440
+ "visual.blocks.15.norm2.weight": "model-00001-of-00007.safetensors",
441
+ "visual.blocks.16.attn.proj.bias": "model-00001-of-00007.safetensors",
442
+ "visual.blocks.16.attn.proj.weight": "model-00001-of-00007.safetensors",
443
+ "visual.blocks.16.attn.qkv.bias": "model-00001-of-00007.safetensors",
444
+ "visual.blocks.16.attn.qkv.weight": "model-00001-of-00007.safetensors",
445
+ "visual.blocks.16.mlp.fc1.bias": "model-00001-of-00007.safetensors",
446
+ "visual.blocks.16.mlp.fc1.weight": "model-00001-of-00007.safetensors",
447
+ "visual.blocks.16.mlp.fc2.bias": "model-00001-of-00007.safetensors",
448
+ "visual.blocks.16.mlp.fc2.weight": "model-00001-of-00007.safetensors",
449
+ "visual.blocks.16.norm1.bias": "model-00001-of-00007.safetensors",
450
+ "visual.blocks.16.norm1.weight": "model-00001-of-00007.safetensors",
451
+ "visual.blocks.16.norm2.bias": "model-00001-of-00007.safetensors",
452
+ "visual.blocks.16.norm2.weight": "model-00001-of-00007.safetensors",
453
+ "visual.blocks.17.attn.proj.bias": "model-00001-of-00007.safetensors",
454
+ "visual.blocks.17.attn.proj.weight": "model-00001-of-00007.safetensors",
455
+ "visual.blocks.17.attn.qkv.bias": "model-00001-of-00007.safetensors",
456
+ "visual.blocks.17.attn.qkv.weight": "model-00001-of-00007.safetensors",
457
+ "visual.blocks.17.mlp.fc1.bias": "model-00001-of-00007.safetensors",
458
+ "visual.blocks.17.mlp.fc1.weight": "model-00001-of-00007.safetensors",
459
+ "visual.blocks.17.mlp.fc2.bias": "model-00001-of-00007.safetensors",
460
+ "visual.blocks.17.mlp.fc2.weight": "model-00001-of-00007.safetensors",
461
+ "visual.blocks.17.norm1.bias": "model-00001-of-00007.safetensors",
462
+ "visual.blocks.17.norm1.weight": "model-00001-of-00007.safetensors",
463
+ "visual.blocks.17.norm2.bias": "model-00001-of-00007.safetensors",
464
+ "visual.blocks.17.norm2.weight": "model-00001-of-00007.safetensors",
465
+ "visual.blocks.18.attn.proj.bias": "model-00001-of-00007.safetensors",
466
+ "visual.blocks.18.attn.proj.weight": "model-00001-of-00007.safetensors",
467
+ "visual.blocks.18.attn.qkv.bias": "model-00001-of-00007.safetensors",
468
+ "visual.blocks.18.attn.qkv.weight": "model-00001-of-00007.safetensors",
469
+ "visual.blocks.18.mlp.fc1.bias": "model-00001-of-00007.safetensors",
470
+ "visual.blocks.18.mlp.fc1.weight": "model-00001-of-00007.safetensors",
471
+ "visual.blocks.18.mlp.fc2.bias": "model-00001-of-00007.safetensors",
472
+ "visual.blocks.18.mlp.fc2.weight": "model-00001-of-00007.safetensors",
473
+ "visual.blocks.18.norm1.bias": "model-00001-of-00007.safetensors",
474
+ "visual.blocks.18.norm1.weight": "model-00001-of-00007.safetensors",
475
+ "visual.blocks.18.norm2.bias": "model-00001-of-00007.safetensors",
476
+ "visual.blocks.18.norm2.weight": "model-00001-of-00007.safetensors",
477
+ "visual.blocks.19.attn.proj.bias": "model-00001-of-00007.safetensors",
478
+ "visual.blocks.19.attn.proj.weight": "model-00001-of-00007.safetensors",
479
+ "visual.blocks.19.attn.qkv.bias": "model-00001-of-00007.safetensors",
480
+ "visual.blocks.19.attn.qkv.weight": "model-00001-of-00007.safetensors",
481
+ "visual.blocks.19.mlp.fc1.bias": "model-00001-of-00007.safetensors",
482
+ "visual.blocks.19.mlp.fc1.weight": "model-00001-of-00007.safetensors",
483
+ "visual.blocks.19.mlp.fc2.bias": "model-00001-of-00007.safetensors",
484
+ "visual.blocks.19.mlp.fc2.weight": "model-00001-of-00007.safetensors",
485
+ "visual.blocks.19.norm1.bias": "model-00001-of-00007.safetensors",
486
+ "visual.blocks.19.norm1.weight": "model-00001-of-00007.safetensors",
487
+ "visual.blocks.19.norm2.bias": "model-00001-of-00007.safetensors",
488
+ "visual.blocks.19.norm2.weight": "model-00001-of-00007.safetensors",
489
+ "visual.blocks.2.attn.proj.bias": "model-00001-of-00007.safetensors",
490
+ "visual.blocks.2.attn.proj.weight": "model-00001-of-00007.safetensors",
491
+ "visual.blocks.2.attn.qkv.bias": "model-00001-of-00007.safetensors",
492
+ "visual.blocks.2.attn.qkv.weight": "model-00001-of-00007.safetensors",
493
+ "visual.blocks.2.mlp.fc1.bias": "model-00001-of-00007.safetensors",
494
+ "visual.blocks.2.mlp.fc1.weight": "model-00001-of-00007.safetensors",
495
+ "visual.blocks.2.mlp.fc2.bias": "model-00001-of-00007.safetensors",
496
+ "visual.blocks.2.mlp.fc2.weight": "model-00001-of-00007.safetensors",
497
+ "visual.blocks.2.norm1.bias": "model-00001-of-00007.safetensors",
498
+ "visual.blocks.2.norm1.weight": "model-00001-of-00007.safetensors",
499
+ "visual.blocks.2.norm2.bias": "model-00001-of-00007.safetensors",
500
+ "visual.blocks.2.norm2.weight": "model-00001-of-00007.safetensors",
501
+ "visual.blocks.20.attn.proj.bias": "model-00001-of-00007.safetensors",
502
+ "visual.blocks.20.attn.proj.weight": "model-00001-of-00007.safetensors",
503
+ "visual.blocks.20.attn.qkv.bias": "model-00001-of-00007.safetensors",
504
+ "visual.blocks.20.attn.qkv.weight": "model-00001-of-00007.safetensors",
505
+ "visual.blocks.20.mlp.fc1.bias": "model-00001-of-00007.safetensors",
506
+ "visual.blocks.20.mlp.fc1.weight": "model-00001-of-00007.safetensors",
507
+ "visual.blocks.20.mlp.fc2.bias": "model-00001-of-00007.safetensors",
508
+ "visual.blocks.20.mlp.fc2.weight": "model-00001-of-00007.safetensors",
509
+ "visual.blocks.20.norm1.bias": "model-00001-of-00007.safetensors",
510
+ "visual.blocks.20.norm1.weight": "model-00001-of-00007.safetensors",
511
+ "visual.blocks.20.norm2.bias": "model-00001-of-00007.safetensors",
512
+ "visual.blocks.20.norm2.weight": "model-00001-of-00007.safetensors",
513
+ "visual.blocks.21.attn.proj.bias": "model-00001-of-00007.safetensors",
514
+ "visual.blocks.21.attn.proj.weight": "model-00001-of-00007.safetensors",
515
+ "visual.blocks.21.attn.qkv.bias": "model-00001-of-00007.safetensors",
516
+ "visual.blocks.21.attn.qkv.weight": "model-00001-of-00007.safetensors",
517
+ "visual.blocks.21.mlp.fc1.bias": "model-00001-of-00007.safetensors",
518
+ "visual.blocks.21.mlp.fc1.weight": "model-00001-of-00007.safetensors",
519
+ "visual.blocks.21.mlp.fc2.bias": "model-00001-of-00007.safetensors",
520
+ "visual.blocks.21.mlp.fc2.weight": "model-00001-of-00007.safetensors",
521
+ "visual.blocks.21.norm1.bias": "model-00001-of-00007.safetensors",
522
+ "visual.blocks.21.norm1.weight": "model-00001-of-00007.safetensors",
523
+ "visual.blocks.21.norm2.bias": "model-00001-of-00007.safetensors",
524
+ "visual.blocks.21.norm2.weight": "model-00001-of-00007.safetensors",
525
+ "visual.blocks.22.attn.proj.bias": "model-00001-of-00007.safetensors",
526
+ "visual.blocks.22.attn.proj.weight": "model-00001-of-00007.safetensors",
527
+ "visual.blocks.22.attn.qkv.bias": "model-00001-of-00007.safetensors",
528
+ "visual.blocks.22.attn.qkv.weight": "model-00001-of-00007.safetensors",
529
+ "visual.blocks.22.mlp.fc1.bias": "model-00001-of-00007.safetensors",
530
+ "visual.blocks.22.mlp.fc1.weight": "model-00001-of-00007.safetensors",
531
+ "visual.blocks.22.mlp.fc2.bias": "model-00001-of-00007.safetensors",
532
+ "visual.blocks.22.mlp.fc2.weight": "model-00001-of-00007.safetensors",
533
+ "visual.blocks.22.norm1.bias": "model-00001-of-00007.safetensors",
534
+ "visual.blocks.22.norm1.weight": "model-00001-of-00007.safetensors",
535
+ "visual.blocks.22.norm2.bias": "model-00001-of-00007.safetensors",
536
+ "visual.blocks.22.norm2.weight": "model-00001-of-00007.safetensors",
537
+ "visual.blocks.23.attn.proj.bias": "model-00001-of-00007.safetensors",
538
+ "visual.blocks.23.attn.proj.weight": "model-00001-of-00007.safetensors",
539
+ "visual.blocks.23.attn.qkv.bias": "model-00001-of-00007.safetensors",
540
+ "visual.blocks.23.attn.qkv.weight": "model-00001-of-00007.safetensors",
541
+ "visual.blocks.23.mlp.fc1.bias": "model-00001-of-00007.safetensors",
542
+ "visual.blocks.23.mlp.fc1.weight": "model-00001-of-00007.safetensors",
543
+ "visual.blocks.23.mlp.fc2.bias": "model-00001-of-00007.safetensors",
544
+ "visual.blocks.23.mlp.fc2.weight": "model-00001-of-00007.safetensors",
545
+ "visual.blocks.23.norm1.bias": "model-00001-of-00007.safetensors",
546
+ "visual.blocks.23.norm1.weight": "model-00001-of-00007.safetensors",
547
+ "visual.blocks.23.norm2.bias": "model-00001-of-00007.safetensors",
548
+ "visual.blocks.23.norm2.weight": "model-00001-of-00007.safetensors",
549
+ "visual.blocks.24.attn.proj.bias": "model-00001-of-00007.safetensors",
550
+ "visual.blocks.24.attn.proj.weight": "model-00001-of-00007.safetensors",
551
+ "visual.blocks.24.attn.qkv.bias": "model-00001-of-00007.safetensors",
552
+ "visual.blocks.24.attn.qkv.weight": "model-00001-of-00007.safetensors",
553
+ "visual.blocks.24.mlp.fc1.bias": "model-00001-of-00007.safetensors",
554
+ "visual.blocks.24.mlp.fc1.weight": "model-00001-of-00007.safetensors",
555
+ "visual.blocks.24.mlp.fc2.bias": "model-00001-of-00007.safetensors",
556
+ "visual.blocks.24.mlp.fc2.weight": "model-00001-of-00007.safetensors",
557
+ "visual.blocks.24.norm1.bias": "model-00001-of-00007.safetensors",
558
+ "visual.blocks.24.norm1.weight": "model-00001-of-00007.safetensors",
559
+ "visual.blocks.24.norm2.bias": "model-00001-of-00007.safetensors",
560
+ "visual.blocks.24.norm2.weight": "model-00001-of-00007.safetensors",
561
+ "visual.blocks.25.attn.proj.bias": "model-00001-of-00007.safetensors",
562
+ "visual.blocks.25.attn.proj.weight": "model-00001-of-00007.safetensors",
563
+ "visual.blocks.25.attn.qkv.bias": "model-00001-of-00007.safetensors",
564
+ "visual.blocks.25.attn.qkv.weight": "model-00001-of-00007.safetensors",
565
+ "visual.blocks.25.mlp.fc1.bias": "model-00001-of-00007.safetensors",
566
+ "visual.blocks.25.mlp.fc1.weight": "model-00001-of-00007.safetensors",
567
+ "visual.blocks.25.mlp.fc2.bias": "model-00001-of-00007.safetensors",
568
+ "visual.blocks.25.mlp.fc2.weight": "model-00001-of-00007.safetensors",
569
+ "visual.blocks.25.norm1.bias": "model-00001-of-00007.safetensors",
570
+ "visual.blocks.25.norm1.weight": "model-00001-of-00007.safetensors",
571
+ "visual.blocks.25.norm2.bias": "model-00001-of-00007.safetensors",
572
+ "visual.blocks.25.norm2.weight": "model-00001-of-00007.safetensors",
573
+ "visual.blocks.26.attn.proj.bias": "model-00001-of-00007.safetensors",
574
+ "visual.blocks.26.attn.proj.weight": "model-00001-of-00007.safetensors",
575
+ "visual.blocks.26.attn.qkv.bias": "model-00001-of-00007.safetensors",
576
+ "visual.blocks.26.attn.qkv.weight": "model-00001-of-00007.safetensors",
577
+ "visual.blocks.26.mlp.fc1.bias": "model-00001-of-00007.safetensors",
578
+ "visual.blocks.26.mlp.fc1.weight": "model-00001-of-00007.safetensors",
579
+ "visual.blocks.26.mlp.fc2.bias": "model-00001-of-00007.safetensors",
580
+ "visual.blocks.26.mlp.fc2.weight": "model-00001-of-00007.safetensors",
581
+ "visual.blocks.26.norm1.bias": "model-00001-of-00007.safetensors",
582
+ "visual.blocks.26.norm1.weight": "model-00001-of-00007.safetensors",
583
+ "visual.blocks.26.norm2.bias": "model-00001-of-00007.safetensors",
584
+ "visual.blocks.26.norm2.weight": "model-00001-of-00007.safetensors",
585
+ "visual.blocks.27.attn.proj.bias": "model-00001-of-00007.safetensors",
586
+ "visual.blocks.27.attn.proj.weight": "model-00001-of-00007.safetensors",
587
+ "visual.blocks.27.attn.qkv.bias": "model-00001-of-00007.safetensors",
588
+ "visual.blocks.27.attn.qkv.weight": "model-00001-of-00007.safetensors",
589
+ "visual.blocks.27.mlp.fc1.bias": "model-00001-of-00007.safetensors",
590
+ "visual.blocks.27.mlp.fc1.weight": "model-00001-of-00007.safetensors",
591
+ "visual.blocks.27.mlp.fc2.bias": "model-00001-of-00007.safetensors",
592
+ "visual.blocks.27.mlp.fc2.weight": "model-00001-of-00007.safetensors",
593
+ "visual.blocks.27.norm1.bias": "model-00001-of-00007.safetensors",
594
+ "visual.blocks.27.norm1.weight": "model-00001-of-00007.safetensors",
595
+ "visual.blocks.27.norm2.bias": "model-00001-of-00007.safetensors",
596
+ "visual.blocks.27.norm2.weight": "model-00001-of-00007.safetensors",
597
+ "visual.blocks.28.attn.proj.bias": "model-00001-of-00007.safetensors",
598
+ "visual.blocks.28.attn.proj.weight": "model-00001-of-00007.safetensors",
599
+ "visual.blocks.28.attn.qkv.bias": "model-00001-of-00007.safetensors",
600
+ "visual.blocks.28.attn.qkv.weight": "model-00001-of-00007.safetensors",
601
+ "visual.blocks.28.mlp.fc1.bias": "model-00001-of-00007.safetensors",
602
+ "visual.blocks.28.mlp.fc1.weight": "model-00001-of-00007.safetensors",
603
+ "visual.blocks.28.mlp.fc2.bias": "model-00001-of-00007.safetensors",
604
+ "visual.blocks.28.mlp.fc2.weight": "model-00001-of-00007.safetensors",
605
+ "visual.blocks.28.norm1.bias": "model-00001-of-00007.safetensors",
606
+ "visual.blocks.28.norm1.weight": "model-00001-of-00007.safetensors",
607
+ "visual.blocks.28.norm2.bias": "model-00001-of-00007.safetensors",
608
+ "visual.blocks.28.norm2.weight": "model-00001-of-00007.safetensors",
609
+ "visual.blocks.29.attn.proj.bias": "model-00001-of-00007.safetensors",
610
+ "visual.blocks.29.attn.proj.weight": "model-00001-of-00007.safetensors",
611
+ "visual.blocks.29.attn.qkv.bias": "model-00001-of-00007.safetensors",
612
+ "visual.blocks.29.attn.qkv.weight": "model-00001-of-00007.safetensors",
613
+ "visual.blocks.29.mlp.fc1.bias": "model-00001-of-00007.safetensors",
614
+ "visual.blocks.29.mlp.fc1.weight": "model-00001-of-00007.safetensors",
615
+ "visual.blocks.29.mlp.fc2.bias": "model-00001-of-00007.safetensors",
616
+ "visual.blocks.29.mlp.fc2.weight": "model-00001-of-00007.safetensors",
617
+ "visual.blocks.29.norm1.bias": "model-00001-of-00007.safetensors",
618
+ "visual.blocks.29.norm1.weight": "model-00001-of-00007.safetensors",
619
+ "visual.blocks.29.norm2.bias": "model-00001-of-00007.safetensors",
620
+ "visual.blocks.29.norm2.weight": "model-00001-of-00007.safetensors",
621
+ "visual.blocks.3.attn.proj.bias": "model-00001-of-00007.safetensors",
622
+ "visual.blocks.3.attn.proj.weight": "model-00001-of-00007.safetensors",
623
+ "visual.blocks.3.attn.qkv.bias": "model-00001-of-00007.safetensors",
624
+ "visual.blocks.3.attn.qkv.weight": "model-00001-of-00007.safetensors",
625
+ "visual.blocks.3.mlp.fc1.bias": "model-00001-of-00007.safetensors",
626
+ "visual.blocks.3.mlp.fc1.weight": "model-00001-of-00007.safetensors",
627
+ "visual.blocks.3.mlp.fc2.bias": "model-00001-of-00007.safetensors",
628
+ "visual.blocks.3.mlp.fc2.weight": "model-00001-of-00007.safetensors",
629
+ "visual.blocks.3.norm1.bias": "model-00001-of-00007.safetensors",
630
+ "visual.blocks.3.norm1.weight": "model-00001-of-00007.safetensors",
631
+ "visual.blocks.3.norm2.bias": "model-00001-of-00007.safetensors",
632
+ "visual.blocks.3.norm2.weight": "model-00001-of-00007.safetensors",
633
+ "visual.blocks.30.attn.proj.bias": "model-00001-of-00007.safetensors",
634
+ "visual.blocks.30.attn.proj.weight": "model-00001-of-00007.safetensors",
635
+ "visual.blocks.30.attn.qkv.bias": "model-00001-of-00007.safetensors",
636
+ "visual.blocks.30.attn.qkv.weight": "model-00001-of-00007.safetensors",
637
+ "visual.blocks.30.mlp.fc1.bias": "model-00001-of-00007.safetensors",
638
+ "visual.blocks.30.mlp.fc1.weight": "model-00001-of-00007.safetensors",
639
+ "visual.blocks.30.mlp.fc2.bias": "model-00001-of-00007.safetensors",
640
+ "visual.blocks.30.mlp.fc2.weight": "model-00001-of-00007.safetensors",
641
+ "visual.blocks.30.norm1.bias": "model-00001-of-00007.safetensors",
642
+ "visual.blocks.30.norm1.weight": "model-00001-of-00007.safetensors",
643
+ "visual.blocks.30.norm2.bias": "model-00001-of-00007.safetensors",
644
+ "visual.blocks.30.norm2.weight": "model-00001-of-00007.safetensors",
645
+ "visual.blocks.31.attn.proj.bias": "model-00001-of-00007.safetensors",
646
+ "visual.blocks.31.attn.proj.weight": "model-00001-of-00007.safetensors",
647
+ "visual.blocks.31.attn.qkv.bias": "model-00001-of-00007.safetensors",
648
+ "visual.blocks.31.attn.qkv.weight": "model-00001-of-00007.safetensors",
649
+ "visual.blocks.31.mlp.fc1.bias": "model-00001-of-00007.safetensors",
650
+ "visual.blocks.31.mlp.fc1.weight": "model-00001-of-00007.safetensors",
651
+ "visual.blocks.31.mlp.fc2.bias": "model-00001-of-00007.safetensors",
652
+ "visual.blocks.31.mlp.fc2.weight": "model-00001-of-00007.safetensors",
653
+ "visual.blocks.31.norm1.bias": "model-00001-of-00007.safetensors",
654
+ "visual.blocks.31.norm1.weight": "model-00001-of-00007.safetensors",
655
+ "visual.blocks.31.norm2.bias": "model-00001-of-00007.safetensors",
656
+ "visual.blocks.31.norm2.weight": "model-00001-of-00007.safetensors",
657
+ "visual.blocks.4.attn.proj.bias": "model-00001-of-00007.safetensors",
658
+ "visual.blocks.4.attn.proj.weight": "model-00001-of-00007.safetensors",
659
+ "visual.blocks.4.attn.qkv.bias": "model-00001-of-00007.safetensors",
660
+ "visual.blocks.4.attn.qkv.weight": "model-00001-of-00007.safetensors",
661
+ "visual.blocks.4.mlp.fc1.bias": "model-00001-of-00007.safetensors",
662
+ "visual.blocks.4.mlp.fc1.weight": "model-00001-of-00007.safetensors",
663
+ "visual.blocks.4.mlp.fc2.bias": "model-00001-of-00007.safetensors",
664
+ "visual.blocks.4.mlp.fc2.weight": "model-00001-of-00007.safetensors",
665
+ "visual.blocks.4.norm1.bias": "model-00001-of-00007.safetensors",
666
+ "visual.blocks.4.norm1.weight": "model-00001-of-00007.safetensors",
667
+ "visual.blocks.4.norm2.bias": "model-00001-of-00007.safetensors",
668
+ "visual.blocks.4.norm2.weight": "model-00001-of-00007.safetensors",
669
+ "visual.blocks.5.attn.proj.bias": "model-00001-of-00007.safetensors",
670
+ "visual.blocks.5.attn.proj.weight": "model-00001-of-00007.safetensors",
671
+ "visual.blocks.5.attn.qkv.bias": "model-00001-of-00007.safetensors",
672
+ "visual.blocks.5.attn.qkv.weight": "model-00001-of-00007.safetensors",
673
+ "visual.blocks.5.mlp.fc1.bias": "model-00001-of-00007.safetensors",
674
+ "visual.blocks.5.mlp.fc1.weight": "model-00001-of-00007.safetensors",
675
+ "visual.blocks.5.mlp.fc2.bias": "model-00001-of-00007.safetensors",
676
+ "visual.blocks.5.mlp.fc2.weight": "model-00001-of-00007.safetensors",
677
+ "visual.blocks.5.norm1.bias": "model-00001-of-00007.safetensors",
678
+ "visual.blocks.5.norm1.weight": "model-00001-of-00007.safetensors",
679
+ "visual.blocks.5.norm2.bias": "model-00001-of-00007.safetensors",
680
+ "visual.blocks.5.norm2.weight": "model-00001-of-00007.safetensors",
681
+ "visual.blocks.6.attn.proj.bias": "model-00001-of-00007.safetensors",
682
+ "visual.blocks.6.attn.proj.weight": "model-00001-of-00007.safetensors",
683
+ "visual.blocks.6.attn.qkv.bias": "model-00001-of-00007.safetensors",
684
+ "visual.blocks.6.attn.qkv.weight": "model-00001-of-00007.safetensors",
685
+ "visual.blocks.6.mlp.fc1.bias": "model-00001-of-00007.safetensors",
686
+ "visual.blocks.6.mlp.fc1.weight": "model-00001-of-00007.safetensors",
687
+ "visual.blocks.6.mlp.fc2.bias": "model-00001-of-00007.safetensors",
688
+ "visual.blocks.6.mlp.fc2.weight": "model-00001-of-00007.safetensors",
689
+ "visual.blocks.6.norm1.bias": "model-00001-of-00007.safetensors",
690
+ "visual.blocks.6.norm1.weight": "model-00001-of-00007.safetensors",
691
+ "visual.blocks.6.norm2.bias": "model-00001-of-00007.safetensors",
692
+ "visual.blocks.6.norm2.weight": "model-00001-of-00007.safetensors",
693
+ "visual.blocks.7.attn.proj.bias": "model-00001-of-00007.safetensors",
694
+ "visual.blocks.7.attn.proj.weight": "model-00001-of-00007.safetensors",
695
+ "visual.blocks.7.attn.qkv.bias": "model-00001-of-00007.safetensors",
696
+ "visual.blocks.7.attn.qkv.weight": "model-00001-of-00007.safetensors",
697
+ "visual.blocks.7.mlp.fc1.bias": "model-00001-of-00007.safetensors",
698
+ "visual.blocks.7.mlp.fc1.weight": "model-00001-of-00007.safetensors",
699
+ "visual.blocks.7.mlp.fc2.bias": "model-00001-of-00007.safetensors",
700
+ "visual.blocks.7.mlp.fc2.weight": "model-00001-of-00007.safetensors",
701
+ "visual.blocks.7.norm1.bias": "model-00001-of-00007.safetensors",
702
+ "visual.blocks.7.norm1.weight": "model-00001-of-00007.safetensors",
703
+ "visual.blocks.7.norm2.bias": "model-00001-of-00007.safetensors",
704
+ "visual.blocks.7.norm2.weight": "model-00001-of-00007.safetensors",
705
+ "visual.blocks.8.attn.proj.bias": "model-00001-of-00007.safetensors",
706
+ "visual.blocks.8.attn.proj.weight": "model-00001-of-00007.safetensors",
707
+ "visual.blocks.8.attn.qkv.bias": "model-00001-of-00007.safetensors",
708
+ "visual.blocks.8.attn.qkv.weight": "model-00001-of-00007.safetensors",
709
+ "visual.blocks.8.mlp.fc1.bias": "model-00001-of-00007.safetensors",
710
+ "visual.blocks.8.mlp.fc1.weight": "model-00001-of-00007.safetensors",
711
+ "visual.blocks.8.mlp.fc2.bias": "model-00001-of-00007.safetensors",
712
+ "visual.blocks.8.mlp.fc2.weight": "model-00001-of-00007.safetensors",
713
+ "visual.blocks.8.norm1.bias": "model-00001-of-00007.safetensors",
714
+ "visual.blocks.8.norm1.weight": "model-00001-of-00007.safetensors",
715
+ "visual.blocks.8.norm2.bias": "model-00001-of-00007.safetensors",
716
+ "visual.blocks.8.norm2.weight": "model-00001-of-00007.safetensors",
717
+ "visual.blocks.9.attn.proj.bias": "model-00001-of-00007.safetensors",
718
+ "visual.blocks.9.attn.proj.weight": "model-00001-of-00007.safetensors",
719
+ "visual.blocks.9.attn.qkv.bias": "model-00001-of-00007.safetensors",
720
+ "visual.blocks.9.attn.qkv.weight": "model-00001-of-00007.safetensors",
721
+ "visual.blocks.9.mlp.fc1.bias": "model-00001-of-00007.safetensors",
722
+ "visual.blocks.9.mlp.fc1.weight": "model-00001-of-00007.safetensors",
723
+ "visual.blocks.9.mlp.fc2.bias": "model-00001-of-00007.safetensors",
724
+ "visual.blocks.9.mlp.fc2.weight": "model-00001-of-00007.safetensors",
725
+ "visual.blocks.9.norm1.bias": "model-00001-of-00007.safetensors",
726
+ "visual.blocks.9.norm1.weight": "model-00001-of-00007.safetensors",
727
+ "visual.blocks.9.norm2.bias": "model-00001-of-00007.safetensors",
728
+ "visual.blocks.9.norm2.weight": "model-00001-of-00007.safetensors",
729
+ "visual.merger.ln_q.bias": "model-00001-of-00007.safetensors",
730
+ "visual.merger.ln_q.weight": "model-00001-of-00007.safetensors",
731
+ "visual.merger.mlp.0.bias": "model-00001-of-00007.safetensors",
732
+ "visual.merger.mlp.0.weight": "model-00001-of-00007.safetensors",
733
+ "visual.merger.mlp.2.bias": "model-00001-of-00007.safetensors",
734
+ "visual.merger.mlp.2.weight": "model-00001-of-00007.safetensors",
735
+ "visual.patch_embed.proj.weight": "model-00001-of-00007.safetensors"
736
+ }
737
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:091aa7594dc2fcfbfa06b9e3c22a5f0562ac14f30375c13af7309407a0e67b8a
3
+ size 11420371
tokenizer_config.json ADDED
@@ -0,0 +1,144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "151643": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "151644": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "151645": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "151646": {
29
+ "content": "<|object_ref_start|>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "151647": {
37
+ "content": "<|object_ref_end|>",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ },
44
+ "151648": {
45
+ "content": "<|box_start|>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": true
51
+ },
52
+ "151649": {
53
+ "content": "<|box_end|>",
54
+ "lstrip": false,
55
+ "normalized": false,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": true
59
+ },
60
+ "151650": {
61
+ "content": "<|quad_start|>",
62
+ "lstrip": false,
63
+ "normalized": false,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": true
67
+ },
68
+ "151651": {
69
+ "content": "<|quad_end|>",
70
+ "lstrip": false,
71
+ "normalized": false,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": true
75
+ },
76
+ "151652": {
77
+ "content": "<|vision_start|>",
78
+ "lstrip": false,
79
+ "normalized": false,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": true
83
+ },
84
+ "151653": {
85
+ "content": "<|vision_end|>",
86
+ "lstrip": false,
87
+ "normalized": false,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": true
91
+ },
92
+ "151654": {
93
+ "content": "<|vision_pad|>",
94
+ "lstrip": false,
95
+ "normalized": false,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": true
99
+ },
100
+ "151655": {
101
+ "content": "<|image_pad|>",
102
+ "lstrip": false,
103
+ "normalized": false,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": true
107
+ },
108
+ "151656": {
109
+ "content": "<|video_pad|>",
110
+ "lstrip": false,
111
+ "normalized": false,
112
+ "rstrip": false,
113
+ "single_word": false,
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
+ "model_max_length": 32768,
138
+ "pad_token": "<|endoftext|>",
139
+ "padding_side": "right",
140
+ "processor_class": "Qwen2VLProcessor",
141
+ "split_special_tokens": false,
142
+ "tokenizer_class": "Qwen2Tokenizer",
143
+ "unk_token": null
144
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff