danielhanchen commited on
Commit
d274363
·
verified ·
1 Parent(s): b060d7e

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* 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
 
 
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
README.md ADDED
@@ -0,0 +1,592 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: vllm
3
+ language:
4
+ - en
5
+ - fr
6
+ - es
7
+ - de
8
+ - it
9
+ - pt
10
+ - nl
11
+ - zh
12
+ - ja
13
+ - ko
14
+ - ar
15
+ license: apache-2.0
16
+ inference: false
17
+ base_model:
18
+ - mistralai/Ministral-3-14B-Reasoning-2512
19
+ extra_gated_description: >-
20
+ If you want to learn more about how we process your personal data, please read
21
+ our <a href="https://mistral.ai/terms/">Privacy Policy</a>.
22
+ tags:
23
+ - mistral-common
24
+ - unsloth
25
+ ---
26
+ <div>
27
+ <p style="margin-top: 0;margin-bottom: 0;">
28
+ <em><a href="https://docs.unsloth.ai/basics/unsloth-dynamic-v2.0-gguf">Unsloth Dynamic 2.0</a> achieves superior accuracy & outperforms other leading quants.</em>
29
+ </p>
30
+ <div style="display: flex; gap: 5px; align-items: center; ">
31
+ <a href="https://github.com/unslothai/unsloth/">
32
+ <img src="https://github.com/unslothai/unsloth/raw/main/images/unsloth%20new%20logo.png" width="133">
33
+ </a>
34
+ <a href="https://discord.gg/unsloth">
35
+ <img src="https://github.com/unslothai/unsloth/raw/main/images/Discord%20button.png" width="173">
36
+ </a>
37
+ <a href="https://docs.unsloth.ai/">
38
+ <img src="https://raw.githubusercontent.com/unslothai/unsloth/refs/heads/main/images/documentation%20green%20button.png" width="143">
39
+ </a>
40
+ </div>
41
+ </div>
42
+
43
+
44
+ # Ministral 3 14B Reasoning 2512
45
+ The largest model in the Ministral 3 family, **Ministral 3 14B** offers frontier capabilities and performance comparable to its larger [Mistral Small 3.2 24B](https://huggingface.co/mistralai/Mistral-Small-3.2-Instruct-2506) counterpart. A powerful and efficient language model with vision capabilities.
46
+
47
+ This model is the reasoning post-trained version, trained for reasoning tasks, making it ideal for math, coding and stem related use cases.
48
+
49
+ The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware. Ministral 3 14B can even be deployed locally, capable of fitting in 32GB of VRAM in BF16, and less than 24GB of RAM/VRAM when quantized.
50
+
51
+ ## Key Features
52
+ Ministral 3 14B consists of two main architectural components:
53
+ - **13.5B Language Model**
54
+ - **0.4B Vision Encoder**
55
+
56
+ The Ministral 3 14B Reasoning model offers the following capabilities:
57
+ - **Vision**: Enables the model to analyze images and provide insights based on visual content, in addition to text.
58
+ - **Multilingual**: Supports dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean, Arabic.
59
+ - **System Prompt**: Maintains strong adherence and support for system prompts.
60
+ - **Agentic**: Offers best-in-class agentic capabilities with native function calling and JSON outputting.
61
+ - **Reasoning**: Excels at complex, multi-step reasoning and dynamic problem-solving.
62
+ - **Edge-Optimized**: Delivers best-in-class performance at a small scale, deployable anywhere.
63
+ - **Apache 2.0 License**: Open-source license allowing usage and modification for both commercial and non-commercial purposes.
64
+ - **Large Context Window**: Supports a 256k context window.
65
+
66
+ ### Use Cases
67
+ Private AI deployments where advanced capabilities meet practical hardware constraints:
68
+ - Private/custom chat and AI assistant deployments in constrained environments
69
+ - Advanced local agentic use cases
70
+ - Fine-tuning and specialization
71
+ - And more...
72
+
73
+ Bringing advanced AI capabilities to most environments.
74
+
75
+ ## Ministral 3 Family
76
+
77
+ | Model Name | Type | Precision | Link |
78
+ |--------------------------------|--------------------|-----------|------------------------------------------------------------------------------------------|
79
+ | Ministral 3 3B Base 2512 | Base pre-trained | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-3B-Base-2512) |
80
+ | Ministral 3 3B Instruct 2512 | Instruct post-trained | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-3B-Instruct-2512) |
81
+ | Ministral 3 3B Reasoning 2512 | Reasoning capable | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-3B-Reasoning-2512) |
82
+ | Ministral 3 8B Base 2512 | Base pre-trained | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-8B-Base-2512) |
83
+ | Ministral 3 8B Instruct 2512 | Instruct post-trained | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-8B-Instruct-2512) |
84
+ | Ministral 3 8B Reasoning 2512 | Reasoning capable | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-8B-Reasoning-2512) |
85
+ | Ministral 3 14B Base 2512 | Base pre-trained** | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-14B-Base-2512) |
86
+ | Ministral 3 14B Instruct 2512 | Instruct post-trained | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512) |
87
+ | **Ministral 3 14B Reasoning 2512** | **Reasoning capable** | **BF16** | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-14B-Reasoning-2512) |
88
+
89
+ Other formats available [here](https://huggingface.co/collections/mistralai/ministral-3-quants).
90
+
91
+ ## Benchmark Results
92
+
93
+ We compare Ministral 3 to similar sized models.
94
+
95
+ ### Reasoning
96
+
97
+ | Model | AIME25 | AIME24 | GPQA Diamond | LiveCodeBench |
98
+ |---------------------------|-------------|-------------|--------------|---------------|
99
+ | **Ministral 3 14B** | <u>0.850</u>| <u>0.898</u>| <u>0.712</u> | <u>0.646</u> |
100
+ | Qwen3-14B (Thinking) | 0.737 | 0.837 | 0.663 | 0.593 |
101
+ | | | | | |
102
+ | **Ministral 3 8B** | 0.787 | <u>0.860</u>| 0.668 | <u>0.616</u> |
103
+ | Qwen3-VL-8B-Thinking | <u>0.798</u>| <u>0.860</u>| <u>0.671</u> | 0.580 |
104
+ | | | | | |
105
+ | **Ministral 3 3B** | <u>0.721</u>| <u>0.775</u>| 0.534 | <u>0.548</u> |
106
+ | Qwen3-VL-4B-Thinking | 0.697 | 0.729 | <u>0.601</u> | 0.513 |
107
+
108
+ ### Instruct
109
+
110
+ | Model | Arena Hard | WildBench | MATH Maj@1 | MM MTBench |
111
+ |---------------------------|-------------|------------|-------------|------------------|
112
+ | **Ministral 3 14B** | <u>0.551</u>| <u>68.5</u>| <u>0.904</u>| <u>8.49</u> |
113
+ | Qwen3 14B (Non-Thinking) | 0.427 | 65.1 | 0.870 | NOT MULTIMODAL |
114
+ | Gemma3-12B-Instruct | 0.436 | 63.2 | 0.854 | 6.70 |
115
+ | | | | | |
116
+ | **Ministral 3 8B** | 0.509 | <u>66.8</u>| 0.876 | <u>8.08</u> |
117
+ | Qwen3-VL-8B-Instruct | <u>0.528</u>| 66.3 | <u>0.946</u>| 8.00 |
118
+ | | | | | |
119
+ | **Ministral 3 3B** | 0.305 | <u>56.8</u>| 0.830 | 7.83 |
120
+ | Qwen3-VL-4B-Instruct | <u>0.438</u>| <u>56.8</u>| <u>0.900</u>| <u>8.01</u> |
121
+ | Qwen3-VL-2B-Instruct | 0.163 | 42.2 | 0.786 | 6.36 |
122
+ | Gemma3-4B-Instruct | 0.318 | 49.1 | 0.759 | 5.23 |
123
+
124
+ ### Base
125
+
126
+ | Model | Multilingual MMLU | MATH CoT 2-Shot | AGIEval 5-shot | MMLU Redux 5-shot | MMLU 5-shot | TriviaQA 5-shot |
127
+ |---------------------|-------------------|-----------------|----------------|-------------------|-------------|-----------------|
128
+ | **Ministral 3 14B** | 0.742 | <u>0.676</u> | 0.648 | 0.820 | 0.794 | 0.749 |
129
+ | Qwen3 14B Base | <u>0.754</u> | 0.620 | <u>0.661</u> | <u>0.837</u> | <u>0.804</u>| 0.703 |
130
+ | Gemma 3 12B Base | 0.690 | 0.487 | 0.587 | 0.766 | 0.745 | <u>0.788</u> |
131
+ | | | | | | | |
132
+ | **Ministral 3 8B** | <u>0.706</u> | <u>0.626</u> | 0.591 | 0.793 | <u>0.761</u>| <u>0.681</u> |
133
+ | Qwen 3 8B Base | 0.700 | 0.576 | <u>0.596</u> | <u>0.794</u> | 0.760 | 0.639 |
134
+ | | | | | | | |
135
+ | **Ministral 3 3B** | 0.652 | <u>0.601</u> | 0.511 | 0.735 | 0.707 | 0.592 |
136
+ | Qwen 3 4B Base | <u>0.677</u> | 0.405 | <u>0.570</u> | <u>0.759</u> | <u>0.713</u>| 0.530 |
137
+ | Gemma 3 4B Base | 0.516 | 0.294 | 0.430 | 0.626 | 0.589 | <u>0.640</u> |
138
+
139
+ ## Usage
140
+
141
+ The model can be used with the following frameworks;
142
+ - [`vllm`](https://github.com/vllm-project/vllm): See [here](#vllm)
143
+ - [`transformers`](https://github.com/huggingface/transformers): See [here](#transformers)
144
+
145
+ ### vLLM
146
+
147
+ We recommend using this model with [vLLM](https://github.com/vllm-project/vllm).
148
+
149
+ #### Installation
150
+
151
+ Make sure to install [`vLLM >= 0.12.0`](https://github.com/vllm-project/vllm/releases/tag/v0.12.0):
152
+
153
+ ```
154
+ pip install vllm --upgrade
155
+ ```
156
+
157
+ Doing so should automatically install [`mistral_common >= 1.8.6`](https://github.com/mistralai/mistral-common/releases/tag/v1.8.6).
158
+
159
+ To check:
160
+ ```
161
+ python -c "import mistral_common; print(mistral_common.__version__)"
162
+ ```
163
+
164
+ You can also make use of a ready-to-go [docker image](https://github.com/vllm-project/vllm/blob/main/Dockerfile) or on the [docker hub](https://hub.docker.com/layers/vllm/vllm-openai/latest/images/sha256-de9032a92ffea7b5c007dad80b38fd44aac11eddc31c435f8e52f3b7404bbf39).
165
+
166
+ #### Serve
167
+
168
+ To fully exploit the `Ministral-3-14B-Reasoning-2512` we recommed using 2xH200 GPUs for deployment due to its large context. However if you don't need a large context, you can fall back to a single GPU.
169
+
170
+ A simple launch command is:
171
+
172
+ ```bash
173
+
174
+ vllm serve mistralai/Ministral-3-14B-Reasoning-2512-FP8 \
175
+ --tensor-parallel-size 2 \
176
+ --enable-auto-tool-choice --tool-call-parser mistral \
177
+ --reasoning-parser mistral
178
+ ```
179
+
180
+ Key parameter notes:
181
+
182
+ * enable-auto-tool-choice: Required when enabling tool usage.
183
+ * tool-call-parser mistral: Required when enabling tool usage.
184
+ * reasoning-parser mistral: Required when enabling reasoning.
185
+
186
+ Additional flags:
187
+
188
+ * You can set `--max-model-len` to preserve memory. By default it is set to `262144` which is quite large but not necessary for most scenarios.
189
+ * You can set `--max-num-batched-tokens` to balance throughput and latency, higher means higher throughput but higher latency.
190
+
191
+ #### Usage of the model
192
+
193
+ Here we asumme that the model `mistralai/Ministral-3-8B-Reasoning-2512` is served and you can ping it to the domain `localhost` with the port `8000` which is the default for vLLM.
194
+
195
+ <details>
196
+ <summary>Vision Reasoning</summary>
197
+
198
+ Let's see if the Ministral 3 model knows when to pick a fight !
199
+
200
+ ```python
201
+ from typing import Any
202
+
203
+ from openai import OpenAI
204
+ from huggingface_hub import hf_hub_download
205
+
206
+ # Modify OpenAI's API key and API base to use vLLM's API server.
207
+ openai_api_key = "EMPTY"
208
+ openai_api_base = "http://localhost:8000/v1"
209
+
210
+ TEMP = 0.7
211
+ TOP_P = 0.95
212
+ MAX_TOK = 262144
213
+ client = OpenAI(
214
+ api_key=openai_api_key,
215
+ base_url=openai_api_base,
216
+ )
217
+
218
+ models = client.models.list()
219
+ model = models.data[0].id
220
+
221
+
222
+ def load_system_prompt(repo_id: str, filename: str) -> dict[str, Any]:
223
+ file_path = hf_hub_download(repo_id=repo_id, filename=filename)
224
+ with open(file_path, "r") as file:
225
+ system_prompt = file.read()
226
+
227
+ index_begin_think = system_prompt.find("[THINK]")
228
+ index_end_think = system_prompt.find("[/THINK]")
229
+
230
+ return {
231
+ "role": "system",
232
+ "content": [
233
+ {"type": "text", "text": system_prompt[:index_begin_think]},
234
+ {
235
+ "type": "thinking",
236
+ "thinking": system_prompt[
237
+ index_begin_think + len("[THINK]") : index_end_think
238
+ ],
239
+ "closed": True,
240
+ },
241
+ {
242
+ "type": "text",
243
+ "text": system_prompt[index_end_think + len("[/THINK]") :],
244
+ },
245
+ ],
246
+ }
247
+
248
+
249
+ SYSTEM_PROMPT = load_system_prompt(model, "SYSTEM_PROMPT.txt")
250
+
251
+ image_url = "https://static.wikia.nocookie.net/essentialsdocs/images/7/70/Battle.png/revision/latest?cb=20220523172438"
252
+
253
+ messages = [
254
+ SYSTEM_PROMPT,
255
+ {
256
+ "role": "user",
257
+ "content": [
258
+ {
259
+ "type": "text",
260
+ "text": "What action do you think I should take in this situation? List all the possible actions and explain why you think they are good or bad.",
261
+ },
262
+ {"type": "image_url", "image_url": {"url": image_url}},
263
+ ],
264
+ },
265
+ ]
266
+
267
+
268
+ stream = client.chat.completions.create(
269
+ model=model,
270
+ messages=messages,
271
+ stream=True,
272
+ temperature=TEMP,
273
+ top_p=TOP_P,
274
+ max_tokens=MAX_TOK,
275
+ )
276
+
277
+ print("client: Start streaming chat completions...:\n")
278
+ printed_reasoning_content = False
279
+ answer = []
280
+
281
+ for chunk in stream:
282
+ reasoning_content = None
283
+ content = None
284
+ # Check the content is reasoning_content or content
285
+ if hasattr(chunk.choices[0].delta, "reasoning_content"):
286
+ reasoning_content = chunk.choices[0].delta.reasoning_content
287
+ if hasattr(chunk.choices[0].delta, "content"):
288
+ content = chunk.choices[0].delta.content
289
+
290
+ if reasoning_content is not None:
291
+ if not printed_reasoning_content:
292
+ printed_reasoning_content = True
293
+ print("Start reasoning:\n", end="", flush=True)
294
+ print(reasoning_content, end="", flush=True)
295
+ elif content is not None:
296
+ # Extract and print the content
297
+ if not reasoning_content and printed_reasoning_content:
298
+ answer.extend(content)
299
+ print(content, end="", flush=True)
300
+
301
+ if answer:
302
+ print("\n\n=============\nAnswer\n=============\n")
303
+ print("".join(answer))
304
+ else:
305
+ print("\n\n=============\nNo Answer\n=============\n")
306
+ print(
307
+ "No answer was generated by the model, probably because the maximum number of tokens was reached."
308
+ )
309
+ ```
310
+
311
+ Now we'll make it compute some maths !
312
+
313
+ ```python
314
+ from typing import Any
315
+
316
+ from openai import OpenAI
317
+ from huggingface_hub import hf_hub_download
318
+
319
+ # Modify OpenAI's API key and API base to use vLLM's API server.
320
+ openai_api_key = "EMPTY"
321
+ openai_api_base = "http://localhost:8000/v1"
322
+
323
+ TEMP = 0.7
324
+ TOP_P = 0.95
325
+ MAX_TOK = 262144
326
+ client = OpenAI(
327
+ api_key=openai_api_key,
328
+ base_url=openai_api_base,
329
+ )
330
+
331
+ models = client.models.list()
332
+ model = models.data[0].id
333
+
334
+
335
+ def load_system_prompt(repo_id: str, filename: str) -> dict[str, Any]:
336
+ file_path = hf_hub_download(repo_id=repo_id, filename=filename)
337
+ with open(file_path, "r") as file:
338
+ system_prompt = file.read()
339
+
340
+ index_begin_think = system_prompt.find("[THINK]")
341
+ index_end_think = system_prompt.find("[/THINK]")
342
+
343
+ return {
344
+ "role": "system",
345
+ "content": [
346
+ {"type": "text", "text": system_prompt[:index_begin_think]},
347
+ {
348
+ "type": "thinking",
349
+ "thinking": system_prompt[
350
+ index_begin_think + len("[THINK]") : index_end_think
351
+ ],
352
+ "closed": True,
353
+ },
354
+ {
355
+ "type": "text",
356
+ "text": system_prompt[index_end_think + len("[/THINK]") :],
357
+ },
358
+ ],
359
+ }
360
+
361
+
362
+ SYSTEM_PROMPT = load_system_prompt(model, "SYSTEM_PROMPT.txt")
363
+
364
+ image_url = "https://i.ytimg.com/vi/5Y3xLHeyKZU/hqdefault.jpg"
365
+
366
+ messages = [
367
+ SYSTEM_PROMPT,
368
+ {
369
+ "role": "user",
370
+ "content": [
371
+ {
372
+ "type": "text",
373
+ "text": "Solve the equations. If they contain only numbers, use your calculator, else only think. Answer in the language of the image.",
374
+ },
375
+ {"type": "image_url", "image_url": {"url": image_url}},
376
+ ],
377
+ },
378
+ ]
379
+
380
+ stream = client.chat.completions.create(
381
+ model=model,
382
+ messages=messages,
383
+ stream=True,
384
+ temperature=TEMP,
385
+ top_p=TOP_P,
386
+ max_tokens=MAX_TOK,
387
+ )
388
+
389
+ print("client: Start streaming chat completions...:\n")
390
+ printed_reasoning_content = False
391
+ answer = []
392
+
393
+ for chunk in stream:
394
+ reasoning_content = None
395
+ content = None
396
+ # Check the content is reasoning_content or content
397
+ if hasattr(chunk.choices[0].delta, "reasoning_content"):
398
+ reasoning_content = chunk.choices[0].delta.reasoning_content
399
+ if hasattr(chunk.choices[0].delta, "content"):
400
+ content = chunk.choices[0].delta.content
401
+
402
+ if reasoning_content is not None:
403
+ if not printed_reasoning_content:
404
+ printed_reasoning_content = True
405
+ print("Start reasoning:\n", end="", flush=True)
406
+ print(reasoning_content, end="", flush=True)
407
+ if content is not None:
408
+ # Extract and print the content
409
+ if not reasoning_content and printed_reasoning_content:
410
+ answer.extend(content)
411
+ print(content, end="", flush=True)
412
+
413
+ if answer:
414
+ print("\n\n=============\nAnswer\n=============\n")
415
+ print("".join(answer))
416
+ else:
417
+ print("\n\n=============\nNo Answer\n=============\n")
418
+ print(
419
+ "No answer was generated by the model, probably because the maximum number of tokens was reached."
420
+ )
421
+ ```
422
+
423
+ </details>
424
+
425
+ <details>
426
+ <summary>Text-Only Request</summary>
427
+
428
+ Let's do more maths and leave it up to the model to figure out how to achieve a result.
429
+
430
+ ```python
431
+ from typing import Any
432
+ from openai import OpenAI
433
+ from huggingface_hub import hf_hub_download
434
+
435
+ # Modify OpenAI's API key and API base to use vLLM's API server.
436
+ openai_api_key = "EMPTY"
437
+ openai_api_base = "http://localhost:8000/v1"
438
+
439
+ TEMP = 0.7
440
+ TOP_P = 0.95
441
+ MAX_TOK = 262144
442
+ client = OpenAI(
443
+ api_key=openai_api_key,
444
+ base_url=openai_api_base,
445
+ )
446
+
447
+ models = client.models.list()
448
+ model = models.data[0].id
449
+
450
+
451
+ def load_system_prompt(repo_id: str, filename: str) -> dict[str, Any]:
452
+ file_path = hf_hub_download(repo_id=repo_id, filename=filename)
453
+ with open(file_path, "r") as file:
454
+ system_prompt = file.read()
455
+
456
+ index_begin_think = system_prompt.find("[THINK]")
457
+ index_end_think = system_prompt.find("[/THINK]")
458
+
459
+ return {
460
+ "role": "system",
461
+ "content": [
462
+ {"type": "text", "text": system_prompt[:index_begin_think]},
463
+ {
464
+ "type": "thinking",
465
+ "thinking": system_prompt[
466
+ index_begin_think + len("[THINK]") : index_end_think
467
+ ],
468
+ "closed": True,
469
+ },
470
+ {
471
+ "type": "text",
472
+ "text": system_prompt[index_end_think + len("[/THINK]") :],
473
+ },
474
+ ],
475
+ }
476
+
477
+
478
+ SYSTEM_PROMPT = load_system_prompt(model, "SYSTEM_PROMPT.txt")
479
+
480
+ query = "Use each number in 2,5,6,3 exactly once, along with any combination of +, -, ×, ÷ (and parentheses for grouping), to make the number 24."
481
+
482
+ messages = [
483
+ SYSTEM_PROMPT,
484
+ {"role": "user", "content": query}
485
+ ]
486
+ stream = client.chat.completions.create(
487
+ model=model,
488
+ messages=messages,
489
+ stream=True,
490
+ temperature=TEMP,
491
+ top_p=TOP_P,
492
+ max_tokens=MAX_TOK,
493
+ )
494
+
495
+ print("client: Start streaming chat completions...:\n")
496
+ printed_reasoning_content = False
497
+ answer = []
498
+
499
+ for chunk in stream:
500
+ reasoning_content = None
501
+ content = None
502
+ # Check the content is reasoning_content or content
503
+ if hasattr(chunk.choices[0].delta, "reasoning_content"):
504
+ reasoning_content = chunk.choices[0].delta.reasoning_content
505
+ if hasattr(chunk.choices[0].delta, "content"):
506
+ content = chunk.choices[0].delta.content
507
+
508
+ if reasoning_content is not None:
509
+ if not printed_reasoning_content:
510
+ printed_reasoning_content = True
511
+ print("Start reasoning:\n", end="", flush=True)
512
+ print(reasoning_content, end="", flush=True)
513
+ if content is not None:
514
+ # Extract and print the content
515
+ if not reasoning_content and printed_reasoning_content:
516
+ answer.extend(content)
517
+ print(content, end="", flush=True)
518
+
519
+ if answer:
520
+ print("\n\n=============\nAnswer\n=============\n")
521
+ print("".join(answer))
522
+ else:
523
+ print("\n\n=============\nNo Answer\n=============\n")
524
+ print("No answer was generated by the model, probably because the maximum number of tokens was reached.")
525
+ ```
526
+
527
+ </details>
528
+
529
+ ### Transformers
530
+
531
+ You can also use Ministral 3 14B Reasoning 2512 with `Transformers` !
532
+
533
+ To make the best use of our model with `Transformers` make sure to have [installed](https://github.com/mistralai/mistral-common) `mistral-common >= 1.8.6` to use our tokenizer.
534
+
535
+ ```bash
536
+ pip install mistral-common --upgrade
537
+ ```
538
+
539
+ Then load our tokenizer along with the model and generate:
540
+
541
+ <details>
542
+ <summary>Python snippet</summary>
543
+
544
+ ```python
545
+ import torch
546
+ from transformers import Mistral3ForConditionalGeneration, MistralCommonBackend
547
+
548
+ model_id = "mistralai/Ministral-3-14B-Reasoning-2512"
549
+
550
+ tokenizer = MistralCommonBackend.from_pretrained(model_id)
551
+ model = Mistral3ForConditionalGeneration.from_pretrained(
552
+ model_id, torch_dtype=torch.bfloat16, device_map="auto"
553
+ )
554
+
555
+ image_url = "https://static.wikia.nocookie.net/essentialsdocs/images/7/70/Battle.png/revision/latest?cb=20220523172438"
556
+
557
+ messages = [
558
+ {
559
+ "role": "user",
560
+ "content": [
561
+ {
562
+ "type": "text",
563
+ "text": "What action do you think I should take in this situation? List all the possible actions and explain why you think they are good or bad.",
564
+ },
565
+ {"type": "image_url", "image_url": {"url": image_url}},
566
+ ],
567
+ },
568
+ ]
569
+
570
+ tokenized = tokenizer.apply_chat_template(messages, return_tensors="pt", return_dict=True)
571
+
572
+ tokenized["input_ids"] = tokenized["input_ids"].to(device="cuda")
573
+ tokenized["pixel_values"] = tokenized["pixel_values"].to(dtype=torch.bfloat16, device="cuda")
574
+ image_sizes = [tokenized["pixel_values"].shape[-2:]]
575
+
576
+ output = model.generate(
577
+ **tokenized,
578
+ image_sizes=image_sizes,
579
+ max_new_tokens=8092,
580
+ )[0]
581
+
582
+ decoded_output = tokenizer.decode(output[len(tokenized["input_ids"][0]):])
583
+ print(decoded_output)
584
+ ```
585
+
586
+ </details>
587
+
588
+ ## License
589
+
590
+ This model is licensed under the [Apache 2.0 License](https://www.apache.org/licenses/LICENSE-2.0.txt).
591
+
592
+ *You must not use this model in a manner that infringes, misappropriates, or otherwise violates any third party’s rights, including intellectual property rights.*
chat_template.jinja ADDED
@@ -0,0 +1,126 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {#- Unsloth template fixes #}
2
+ {#- Default system message if no system prompt is passed. #}
3
+ {%- set default_system_message = '# HOW YOU SHOULD THINK AND ANSWER\n\nFirst draft your thinking process (inner monologue) until you arrive at a response. Format your response using Markdown, and use LaTeX for any mathematical equations. Write both your thoughts and the response in the same language as the input.\n\nYour thinking process must follow the template below:[THINK]Your thoughts or/and draft, like working through an exercise on scratch paper. Be as casual and as long as you want until you are confident to generate the response to the user.[/THINK]Here, provide a self-contained response.' %}
4
+
5
+ {#- Begin of sequence token. #}
6
+ {{- bos_token }}
7
+
8
+ {#- Handle system prompt if it exists. #}
9
+ {#- System prompt supports text content or text and thinking chunks. #}
10
+ {%- if messages[0]['role'] == 'system' %}
11
+ {{- '[SYSTEM_PROMPT]' -}}
12
+ {%- if messages[0]['content'] is string %}
13
+ {{- messages[0]['content'] -}}
14
+ {%- else %}
15
+ {%- for block in messages[0]['content'] %}
16
+ {%- if block['type'] == 'text' %}
17
+ {{- block['text'] }}
18
+ {%- elif block['type'] == 'thinking' %}
19
+ {{- '[THINK]' + block['thinking'] + '[/THINK]' }}
20
+ {%- else %}
21
+ {{- raise_exception('Only text and thinking chunks are supported in system message contents.') }}
22
+ {%- endif %}
23
+ {%- endfor %}
24
+ {%- endif %}
25
+ {{- '[/SYSTEM_PROMPT]' -}}
26
+ {%- set loop_messages = messages[1:] %}
27
+ {%- else %}
28
+ {%- set loop_messages = messages %}
29
+ {%- if default_system_message != '' %}
30
+ {{- '[SYSTEM_PROMPT]' + default_system_message + '[/SYSTEM_PROMPT]' }}
31
+ {%- endif %}
32
+ {%- endif %}
33
+
34
+
35
+ {#- Tools definition #}
36
+ {%- set tools_definition = '' %}
37
+ {%- set has_tools = false %}
38
+ {%- if tools is defined and tools is not none and tools|length > 0 %}
39
+ {%- set has_tools = true %}
40
+ {%- set tools_definition = '[AVAILABLE_TOOLS]' + (tools| tojson) + '[/AVAILABLE_TOOLS]' %}
41
+ {{- tools_definition }}
42
+ {%- endif %}
43
+
44
+ {#- Checks for alternating user/assistant messages. #}
45
+ {%- set ns = namespace(index=0) %}
46
+ {%- for message in loop_messages %}
47
+ {%- if message.role == 'user' or (message.role == 'assistant' and (message.tool_calls is not defined or message.tool_calls is none or message.tool_calls | length == 0)) %}
48
+ {%- if (message['role'] == 'user') != (ns.index % 2 == 0) %}
49
+ {{- raise_exception('After the optional system message, conversation roles must alternate user and assistant roles except for tool calls and results.') }}
50
+ {%- endif %}
51
+ {%- set ns.index = ns.index + 1 %}
52
+ {%- endif %}
53
+ {%- endfor %}
54
+
55
+ {#- Handle conversation messages. #}
56
+ {%- for message in loop_messages %}
57
+
58
+ {#- User messages supports text content or text and image chunks. #}
59
+ {%- if message['role'] == 'user' %}
60
+ {%- if message['content'] is string %}
61
+ {{- '[INST]' + message['content'] + '[/INST]' }}
62
+ {%- elif message['content'] | length > 0 %}
63
+ {{- '[INST]' }}
64
+ {%- if message['content'] | length == 2 %}
65
+ {%- set blocks = message['content'] | sort(attribute='type') %}
66
+ {%- else %}
67
+ {%- set blocks = message['content'] %}
68
+ {%- endif %}
69
+ {%- for block in blocks %}
70
+ {%- if block['type'] == 'text' %}
71
+ {{- block['text'] }}
72
+ {%- elif block['type'] in ['image', 'image_url'] %}
73
+ {{- '[IMG]' }}
74
+ {%- else %}
75
+ {{- raise_exception('Only text, image and image_url chunks are supported in user message content.') }}
76
+ {%- endif %}
77
+ {%- endfor %}
78
+ {{- '[/INST]' }}
79
+ {%- else %}
80
+ {{- raise_exception('User message must have a string or a list of chunks in content') }}
81
+ {%- endif %}
82
+
83
+ {#- Assistant messages supports text content or text, image and thinking chunks. #}
84
+ {%- elif message['role'] == 'assistant' %}
85
+
86
+ {%- if message['content'] is string and message['content'] != '' %}
87
+ {{- message['content'] }}
88
+ {%- elif message['content'] is iterable and message['content'] | length > 0 %}
89
+ {%- for block in message['content'] %}
90
+ {%- if block['type'] == 'text' %}
91
+ {{- block['text'] }}
92
+ {%- elif block['type'] == 'thinking' %}
93
+ {{- '[THINK]' + block['thinking'] + '[/THINK]' }}
94
+ {%- else %}
95
+ {{- raise_exception('Only text and thinking chunks are supported in assistant message contents.') }}
96
+ {%- endif %}
97
+ {%- endfor %}
98
+ {%- endif %}
99
+
100
+ {%- if message['tool_calls'] is defined and message['tool_calls'] is not none and message['tool_calls']|length > 0 %}
101
+ {%- for tool in message['tool_calls'] %}
102
+ {{- '[TOOL_CALLS]' }}
103
+ {%- set name = tool['function']['name'] %}
104
+ {%- set arguments = tool['function']['arguments'] %}
105
+ {%- if arguments is not string %}
106
+ {%- set arguments = arguments|tojson|safe %}
107
+ {%- elif arguments == '' %}
108
+ {%- set arguments = '{}' %}
109
+ {%- endif %}
110
+ {{- name + '[ARGS]' + arguments }}
111
+ {%- endfor %}
112
+ {%- endif %}
113
+
114
+ {{- eos_token }}
115
+
116
+ {#- Tool messages only supports text content. #}
117
+ {%- elif message['role'] == 'tool' %}
118
+ {{- '[TOOL_RESULTS]' + message['content']|string + '[/TOOL_RESULTS]' }}
119
+
120
+ {#- Raise exception for unsupported roles. #}
121
+ {%- else %}
122
+ {{- raise_exception('Only user, assistant and tool roles are supported, got ' + message['role'] + '.') }}
123
+ {%- endif %}
124
+ {%- endfor %}
125
+
126
+ {#- Copyright 2025-present Unsloth. Apache 2.0 License. #}
config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Mistral3ForConditionalGeneration"
4
+ ],
5
+ "bos_token_id": 1,
6
+ "torch_dtype": "bfloat16",
7
+ "eos_token_id": 2,
8
+ "image_token_index": 10,
9
+ "model_type": "mistral3",
10
+ "multimodal_projector_bias": false,
11
+ "pad_token_id": 11,
12
+ "projector_hidden_act": "gelu",
13
+ "spatial_merge_size": 2,
14
+ "text_config": {
15
+ "attention_dropout": 0.0,
16
+ "torch_dtype": "bfloat16",
17
+ "head_dim": 128,
18
+ "hidden_act": "silu",
19
+ "hidden_size": 5120,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 16384,
22
+ "max_position_embeddings": 262144,
23
+ "model_type": "ministral3",
24
+ "num_attention_heads": 32,
25
+ "num_hidden_layers": 40,
26
+ "num_key_value_heads": 8,
27
+ "rms_norm_eps": 1e-05,
28
+ "rope_parameters": {
29
+ "beta_fast": 32.0,
30
+ "beta_slow": 1.0,
31
+ "factor": 16.0,
32
+ "llama_4_scaling_beta": 0.1,
33
+ "mscale": 1.0,
34
+ "mscale_all_dim": 1.0,
35
+ "original_max_position_embeddings": 16384,
36
+ "rope_theta": 1000000000.0,
37
+ "rope_type": "yarn",
38
+ "type": "yarn"
39
+ },
40
+ "sliding_window": null,
41
+ "use_cache": true,
42
+ "vocab_size": 131072
43
+ },
44
+ "transformers_version": "5.0.0.dev0",
45
+ "unsloth_fixed": true,
46
+ "vision_config": {
47
+ "attention_dropout": 0.0,
48
+ "torch_dtype": "bfloat16",
49
+ "head_dim": 64,
50
+ "hidden_act": "silu",
51
+ "hidden_size": 1024,
52
+ "image_size": 1540,
53
+ "initializer_range": 0.02,
54
+ "intermediate_size": 4096,
55
+ "model_type": "pixtral",
56
+ "num_attention_heads": 16,
57
+ "num_channels": 3,
58
+ "num_hidden_layers": 24,
59
+ "patch_size": 14,
60
+ "rope_parameters": {
61
+ "rope_theta": 10000.0,
62
+ "rope_type": "default"
63
+ }
64
+ },
65
+ "vision_feature_layer": -1
66
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 1,
3
+ "eos_token_id": 2,
4
+ "max_length": 262144,
5
+ "pad_token_id": 11,
6
+ "transformers_version": "5.0.0.dev0"
7
+ }
model-00001-of-00006.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:163fef681d41683f2ac17ae56f3b6935188d593d252389fffef11ce148fde021
3
+ size 4925537056
model-00002-of-00006.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1bbf71635f7242e22752cee209350ff0925b447708d3fbf184c1725bca481230
3
+ size 4865565920
model-00003-of-00006.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81eec09dad5d0ed898fd60cfce3ae35a487bafb805d192b40d4b9724d6c330b3
3
+ size 4865565968
model-00004-of-00006.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:648838faaaca188c26901ea32a1ea60c82c7ded26f25098b0b6acda47358bf89
3
+ size 4865565968
model-00005-of-00006.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:142478cca6ed3f38a2c9e43d03bc2a3ba84a9d863078e3adbe294f6392aace51
3
+ size 4865565968
model-00006-of-00006.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c87c0c31b38ee58428ccd217c3579be969e684b3e2058b23a6232e7e208eda87
3
+ size 3502340328
model.safetensors.index.json ADDED
@@ -0,0 +1,593 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_parameters": 13945031680,
4
+ "total_size": 27890063360
5
+ },
6
+ "weight_map": {
7
+ "language_model.lm_head.weight": "model-00006-of-00006.safetensors",
8
+ "language_model.model.embed_tokens.weight": "model-00001-of-00006.safetensors",
9
+ "language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00006.safetensors",
10
+ "language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
11
+ "language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
12
+ "language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
13
+ "language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
14
+ "language_model.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
15
+ "language_model.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
16
+ "language_model.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
17
+ "language_model.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
18
+ "language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00006.safetensors",
19
+ "language_model.model.layers.1.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
20
+ "language_model.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
21
+ "language_model.model.layers.1.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
22
+ "language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
23
+ "language_model.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
24
+ "language_model.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
25
+ "language_model.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
26
+ "language_model.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
27
+ "language_model.model.layers.10.input_layernorm.weight": "model-00002-of-00006.safetensors",
28
+ "language_model.model.layers.10.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
29
+ "language_model.model.layers.10.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
30
+ "language_model.model.layers.10.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
31
+ "language_model.model.layers.10.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
32
+ "language_model.model.layers.10.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
33
+ "language_model.model.layers.10.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
34
+ "language_model.model.layers.10.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
35
+ "language_model.model.layers.10.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
36
+ "language_model.model.layers.11.input_layernorm.weight": "model-00002-of-00006.safetensors",
37
+ "language_model.model.layers.11.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
38
+ "language_model.model.layers.11.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
39
+ "language_model.model.layers.11.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
40
+ "language_model.model.layers.11.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
41
+ "language_model.model.layers.11.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
42
+ "language_model.model.layers.11.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
43
+ "language_model.model.layers.11.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
44
+ "language_model.model.layers.11.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
45
+ "language_model.model.layers.12.input_layernorm.weight": "model-00003-of-00006.safetensors",
46
+ "language_model.model.layers.12.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
47
+ "language_model.model.layers.12.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
48
+ "language_model.model.layers.12.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
49
+ "language_model.model.layers.12.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
50
+ "language_model.model.layers.12.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
51
+ "language_model.model.layers.12.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
52
+ "language_model.model.layers.12.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
53
+ "language_model.model.layers.12.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
54
+ "language_model.model.layers.13.input_layernorm.weight": "model-00003-of-00006.safetensors",
55
+ "language_model.model.layers.13.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
56
+ "language_model.model.layers.13.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
57
+ "language_model.model.layers.13.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
58
+ "language_model.model.layers.13.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
59
+ "language_model.model.layers.13.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
60
+ "language_model.model.layers.13.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
61
+ "language_model.model.layers.13.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
62
+ "language_model.model.layers.13.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
63
+ "language_model.model.layers.14.input_layernorm.weight": "model-00003-of-00006.safetensors",
64
+ "language_model.model.layers.14.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
65
+ "language_model.model.layers.14.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
66
+ "language_model.model.layers.14.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
67
+ "language_model.model.layers.14.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
68
+ "language_model.model.layers.14.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
69
+ "language_model.model.layers.14.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
70
+ "language_model.model.layers.14.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
71
+ "language_model.model.layers.14.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
72
+ "language_model.model.layers.15.input_layernorm.weight": "model-00003-of-00006.safetensors",
73
+ "language_model.model.layers.15.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
74
+ "language_model.model.layers.15.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
75
+ "language_model.model.layers.15.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
76
+ "language_model.model.layers.15.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
77
+ "language_model.model.layers.15.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
78
+ "language_model.model.layers.15.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
79
+ "language_model.model.layers.15.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
80
+ "language_model.model.layers.15.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
81
+ "language_model.model.layers.16.input_layernorm.weight": "model-00003-of-00006.safetensors",
82
+ "language_model.model.layers.16.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
83
+ "language_model.model.layers.16.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
84
+ "language_model.model.layers.16.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
85
+ "language_model.model.layers.16.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
86
+ "language_model.model.layers.16.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
87
+ "language_model.model.layers.16.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
88
+ "language_model.model.layers.16.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
89
+ "language_model.model.layers.16.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
90
+ "language_model.model.layers.17.input_layernorm.weight": "model-00003-of-00006.safetensors",
91
+ "language_model.model.layers.17.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
92
+ "language_model.model.layers.17.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
93
+ "language_model.model.layers.17.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
94
+ "language_model.model.layers.17.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
95
+ "language_model.model.layers.17.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
96
+ "language_model.model.layers.17.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
97
+ "language_model.model.layers.17.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
98
+ "language_model.model.layers.17.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
99
+ "language_model.model.layers.18.input_layernorm.weight": "model-00003-of-00006.safetensors",
100
+ "language_model.model.layers.18.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
101
+ "language_model.model.layers.18.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
102
+ "language_model.model.layers.18.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
103
+ "language_model.model.layers.18.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
104
+ "language_model.model.layers.18.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
105
+ "language_model.model.layers.18.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
106
+ "language_model.model.layers.18.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
107
+ "language_model.model.layers.18.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
108
+ "language_model.model.layers.19.input_layernorm.weight": "model-00003-of-00006.safetensors",
109
+ "language_model.model.layers.19.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
110
+ "language_model.model.layers.19.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
111
+ "language_model.model.layers.19.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
112
+ "language_model.model.layers.19.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
113
+ "language_model.model.layers.19.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
114
+ "language_model.model.layers.19.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
115
+ "language_model.model.layers.19.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
116
+ "language_model.model.layers.19.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
117
+ "language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00006.safetensors",
118
+ "language_model.model.layers.2.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
119
+ "language_model.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
120
+ "language_model.model.layers.2.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
121
+ "language_model.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
122
+ "language_model.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
123
+ "language_model.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
124
+ "language_model.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
125
+ "language_model.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
126
+ "language_model.model.layers.20.input_layernorm.weight": "model-00004-of-00006.safetensors",
127
+ "language_model.model.layers.20.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
128
+ "language_model.model.layers.20.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
129
+ "language_model.model.layers.20.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
130
+ "language_model.model.layers.20.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
131
+ "language_model.model.layers.20.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
132
+ "language_model.model.layers.20.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
133
+ "language_model.model.layers.20.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
134
+ "language_model.model.layers.20.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
135
+ "language_model.model.layers.21.input_layernorm.weight": "model-00004-of-00006.safetensors",
136
+ "language_model.model.layers.21.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
137
+ "language_model.model.layers.21.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
138
+ "language_model.model.layers.21.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
139
+ "language_model.model.layers.21.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
140
+ "language_model.model.layers.21.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
141
+ "language_model.model.layers.21.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
142
+ "language_model.model.layers.21.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
143
+ "language_model.model.layers.21.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
144
+ "language_model.model.layers.22.input_layernorm.weight": "model-00004-of-00006.safetensors",
145
+ "language_model.model.layers.22.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
146
+ "language_model.model.layers.22.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
147
+ "language_model.model.layers.22.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
148
+ "language_model.model.layers.22.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
149
+ "language_model.model.layers.22.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
150
+ "language_model.model.layers.22.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
151
+ "language_model.model.layers.22.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
152
+ "language_model.model.layers.22.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
153
+ "language_model.model.layers.23.input_layernorm.weight": "model-00004-of-00006.safetensors",
154
+ "language_model.model.layers.23.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
155
+ "language_model.model.layers.23.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
156
+ "language_model.model.layers.23.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
157
+ "language_model.model.layers.23.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
158
+ "language_model.model.layers.23.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
159
+ "language_model.model.layers.23.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
160
+ "language_model.model.layers.23.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
161
+ "language_model.model.layers.23.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
162
+ "language_model.model.layers.24.input_layernorm.weight": "model-00004-of-00006.safetensors",
163
+ "language_model.model.layers.24.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
164
+ "language_model.model.layers.24.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
165
+ "language_model.model.layers.24.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
166
+ "language_model.model.layers.24.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
167
+ "language_model.model.layers.24.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
168
+ "language_model.model.layers.24.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
169
+ "language_model.model.layers.24.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
170
+ "language_model.model.layers.24.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
171
+ "language_model.model.layers.25.input_layernorm.weight": "model-00004-of-00006.safetensors",
172
+ "language_model.model.layers.25.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
173
+ "language_model.model.layers.25.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
174
+ "language_model.model.layers.25.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
175
+ "language_model.model.layers.25.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
176
+ "language_model.model.layers.25.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
177
+ "language_model.model.layers.25.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
178
+ "language_model.model.layers.25.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
179
+ "language_model.model.layers.25.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
180
+ "language_model.model.layers.26.input_layernorm.weight": "model-00004-of-00006.safetensors",
181
+ "language_model.model.layers.26.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
182
+ "language_model.model.layers.26.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
183
+ "language_model.model.layers.26.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
184
+ "language_model.model.layers.26.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
185
+ "language_model.model.layers.26.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
186
+ "language_model.model.layers.26.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
187
+ "language_model.model.layers.26.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
188
+ "language_model.model.layers.26.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
189
+ "language_model.model.layers.27.input_layernorm.weight": "model-00004-of-00006.safetensors",
190
+ "language_model.model.layers.27.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
191
+ "language_model.model.layers.27.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
192
+ "language_model.model.layers.27.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
193
+ "language_model.model.layers.27.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
194
+ "language_model.model.layers.27.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
195
+ "language_model.model.layers.27.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
196
+ "language_model.model.layers.27.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
197
+ "language_model.model.layers.27.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
198
+ "language_model.model.layers.28.input_layernorm.weight": "model-00005-of-00006.safetensors",
199
+ "language_model.model.layers.28.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
200
+ "language_model.model.layers.28.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
201
+ "language_model.model.layers.28.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
202
+ "language_model.model.layers.28.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
203
+ "language_model.model.layers.28.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
204
+ "language_model.model.layers.28.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
205
+ "language_model.model.layers.28.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
206
+ "language_model.model.layers.28.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
207
+ "language_model.model.layers.29.input_layernorm.weight": "model-00005-of-00006.safetensors",
208
+ "language_model.model.layers.29.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
209
+ "language_model.model.layers.29.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
210
+ "language_model.model.layers.29.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
211
+ "language_model.model.layers.29.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
212
+ "language_model.model.layers.29.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
213
+ "language_model.model.layers.29.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
214
+ "language_model.model.layers.29.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
215
+ "language_model.model.layers.29.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
216
+ "language_model.model.layers.3.input_layernorm.weight": "model-00001-of-00006.safetensors",
217
+ "language_model.model.layers.3.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
218
+ "language_model.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
219
+ "language_model.model.layers.3.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
220
+ "language_model.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
221
+ "language_model.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
222
+ "language_model.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
223
+ "language_model.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
224
+ "language_model.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
225
+ "language_model.model.layers.30.input_layernorm.weight": "model-00005-of-00006.safetensors",
226
+ "language_model.model.layers.30.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
227
+ "language_model.model.layers.30.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
228
+ "language_model.model.layers.30.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
229
+ "language_model.model.layers.30.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
230
+ "language_model.model.layers.30.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
231
+ "language_model.model.layers.30.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
232
+ "language_model.model.layers.30.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
233
+ "language_model.model.layers.30.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
234
+ "language_model.model.layers.31.input_layernorm.weight": "model-00005-of-00006.safetensors",
235
+ "language_model.model.layers.31.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
236
+ "language_model.model.layers.31.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
237
+ "language_model.model.layers.31.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
238
+ "language_model.model.layers.31.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
239
+ "language_model.model.layers.31.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
240
+ "language_model.model.layers.31.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
241
+ "language_model.model.layers.31.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
242
+ "language_model.model.layers.31.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
243
+ "language_model.model.layers.32.input_layernorm.weight": "model-00005-of-00006.safetensors",
244
+ "language_model.model.layers.32.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
245
+ "language_model.model.layers.32.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
246
+ "language_model.model.layers.32.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
247
+ "language_model.model.layers.32.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
248
+ "language_model.model.layers.32.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
249
+ "language_model.model.layers.32.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
250
+ "language_model.model.layers.32.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
251
+ "language_model.model.layers.32.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
252
+ "language_model.model.layers.33.input_layernorm.weight": "model-00005-of-00006.safetensors",
253
+ "language_model.model.layers.33.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
254
+ "language_model.model.layers.33.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
255
+ "language_model.model.layers.33.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
256
+ "language_model.model.layers.33.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
257
+ "language_model.model.layers.33.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
258
+ "language_model.model.layers.33.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
259
+ "language_model.model.layers.33.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
260
+ "language_model.model.layers.33.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
261
+ "language_model.model.layers.34.input_layernorm.weight": "model-00005-of-00006.safetensors",
262
+ "language_model.model.layers.34.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
263
+ "language_model.model.layers.34.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
264
+ "language_model.model.layers.34.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
265
+ "language_model.model.layers.34.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
266
+ "language_model.model.layers.34.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
267
+ "language_model.model.layers.34.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
268
+ "language_model.model.layers.34.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
269
+ "language_model.model.layers.34.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
270
+ "language_model.model.layers.35.input_layernorm.weight": "model-00005-of-00006.safetensors",
271
+ "language_model.model.layers.35.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
272
+ "language_model.model.layers.35.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
273
+ "language_model.model.layers.35.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
274
+ "language_model.model.layers.35.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
275
+ "language_model.model.layers.35.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
276
+ "language_model.model.layers.35.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
277
+ "language_model.model.layers.35.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
278
+ "language_model.model.layers.35.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
279
+ "language_model.model.layers.36.input_layernorm.weight": "model-00006-of-00006.safetensors",
280
+ "language_model.model.layers.36.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
281
+ "language_model.model.layers.36.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
282
+ "language_model.model.layers.36.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
283
+ "language_model.model.layers.36.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
284
+ "language_model.model.layers.36.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
285
+ "language_model.model.layers.36.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
286
+ "language_model.model.layers.36.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
287
+ "language_model.model.layers.36.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
288
+ "language_model.model.layers.37.input_layernorm.weight": "model-00006-of-00006.safetensors",
289
+ "language_model.model.layers.37.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
290
+ "language_model.model.layers.37.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
291
+ "language_model.model.layers.37.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
292
+ "language_model.model.layers.37.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
293
+ "language_model.model.layers.37.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
294
+ "language_model.model.layers.37.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
295
+ "language_model.model.layers.37.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
296
+ "language_model.model.layers.37.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
297
+ "language_model.model.layers.38.input_layernorm.weight": "model-00006-of-00006.safetensors",
298
+ "language_model.model.layers.38.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
299
+ "language_model.model.layers.38.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
300
+ "language_model.model.layers.38.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
301
+ "language_model.model.layers.38.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
302
+ "language_model.model.layers.38.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
303
+ "language_model.model.layers.38.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
304
+ "language_model.model.layers.38.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
305
+ "language_model.model.layers.38.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
306
+ "language_model.model.layers.39.input_layernorm.weight": "model-00006-of-00006.safetensors",
307
+ "language_model.model.layers.39.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
308
+ "language_model.model.layers.39.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
309
+ "language_model.model.layers.39.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
310
+ "language_model.model.layers.39.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
311
+ "language_model.model.layers.39.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
312
+ "language_model.model.layers.39.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
313
+ "language_model.model.layers.39.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
314
+ "language_model.model.layers.39.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
315
+ "language_model.model.layers.4.input_layernorm.weight": "model-00002-of-00006.safetensors",
316
+ "language_model.model.layers.4.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
317
+ "language_model.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
318
+ "language_model.model.layers.4.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
319
+ "language_model.model.layers.4.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
320
+ "language_model.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
321
+ "language_model.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
322
+ "language_model.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
323
+ "language_model.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
324
+ "language_model.model.layers.5.input_layernorm.weight": "model-00002-of-00006.safetensors",
325
+ "language_model.model.layers.5.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
326
+ "language_model.model.layers.5.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
327
+ "language_model.model.layers.5.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
328
+ "language_model.model.layers.5.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
329
+ "language_model.model.layers.5.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
330
+ "language_model.model.layers.5.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
331
+ "language_model.model.layers.5.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
332
+ "language_model.model.layers.5.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
333
+ "language_model.model.layers.6.input_layernorm.weight": "model-00002-of-00006.safetensors",
334
+ "language_model.model.layers.6.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
335
+ "language_model.model.layers.6.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
336
+ "language_model.model.layers.6.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
337
+ "language_model.model.layers.6.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
338
+ "language_model.model.layers.6.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
339
+ "language_model.model.layers.6.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
340
+ "language_model.model.layers.6.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
341
+ "language_model.model.layers.6.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
342
+ "language_model.model.layers.7.input_layernorm.weight": "model-00002-of-00006.safetensors",
343
+ "language_model.model.layers.7.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
344
+ "language_model.model.layers.7.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
345
+ "language_model.model.layers.7.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
346
+ "language_model.model.layers.7.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
347
+ "language_model.model.layers.7.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
348
+ "language_model.model.layers.7.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
349
+ "language_model.model.layers.7.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
350
+ "language_model.model.layers.7.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
351
+ "language_model.model.layers.8.input_layernorm.weight": "model-00002-of-00006.safetensors",
352
+ "language_model.model.layers.8.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
353
+ "language_model.model.layers.8.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
354
+ "language_model.model.layers.8.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
355
+ "language_model.model.layers.8.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
356
+ "language_model.model.layers.8.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
357
+ "language_model.model.layers.8.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
358
+ "language_model.model.layers.8.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
359
+ "language_model.model.layers.8.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
360
+ "language_model.model.layers.9.input_layernorm.weight": "model-00002-of-00006.safetensors",
361
+ "language_model.model.layers.9.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
362
+ "language_model.model.layers.9.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
363
+ "language_model.model.layers.9.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
364
+ "language_model.model.layers.9.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
365
+ "language_model.model.layers.9.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
366
+ "language_model.model.layers.9.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
367
+ "language_model.model.layers.9.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
368
+ "language_model.model.layers.9.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
369
+ "language_model.model.norm.weight": "model-00006-of-00006.safetensors",
370
+ "multi_modal_projector.linear_1.weight": "model-00001-of-00006.safetensors",
371
+ "multi_modal_projector.linear_2.weight": "model-00001-of-00006.safetensors",
372
+ "multi_modal_projector.norm.weight": "model-00001-of-00006.safetensors",
373
+ "multi_modal_projector.patch_merger.merging_layer.weight": "model-00001-of-00006.safetensors",
374
+ "vision_tower.ln_pre.weight": "model-00001-of-00006.safetensors",
375
+ "vision_tower.patch_conv.weight": "model-00001-of-00006.safetensors",
376
+ "vision_tower.transformer.layers.0.attention.k_proj.weight": "model-00001-of-00006.safetensors",
377
+ "vision_tower.transformer.layers.0.attention.o_proj.weight": "model-00001-of-00006.safetensors",
378
+ "vision_tower.transformer.layers.0.attention.q_proj.weight": "model-00001-of-00006.safetensors",
379
+ "vision_tower.transformer.layers.0.attention.v_proj.weight": "model-00001-of-00006.safetensors",
380
+ "vision_tower.transformer.layers.0.attention_norm.weight": "model-00001-of-00006.safetensors",
381
+ "vision_tower.transformer.layers.0.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
382
+ "vision_tower.transformer.layers.0.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
383
+ "vision_tower.transformer.layers.0.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
384
+ "vision_tower.transformer.layers.0.ffn_norm.weight": "model-00001-of-00006.safetensors",
385
+ "vision_tower.transformer.layers.1.attention.k_proj.weight": "model-00001-of-00006.safetensors",
386
+ "vision_tower.transformer.layers.1.attention.o_proj.weight": "model-00001-of-00006.safetensors",
387
+ "vision_tower.transformer.layers.1.attention.q_proj.weight": "model-00001-of-00006.safetensors",
388
+ "vision_tower.transformer.layers.1.attention.v_proj.weight": "model-00001-of-00006.safetensors",
389
+ "vision_tower.transformer.layers.1.attention_norm.weight": "model-00001-of-00006.safetensors",
390
+ "vision_tower.transformer.layers.1.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
391
+ "vision_tower.transformer.layers.1.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
392
+ "vision_tower.transformer.layers.1.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
393
+ "vision_tower.transformer.layers.1.ffn_norm.weight": "model-00001-of-00006.safetensors",
394
+ "vision_tower.transformer.layers.10.attention.k_proj.weight": "model-00001-of-00006.safetensors",
395
+ "vision_tower.transformer.layers.10.attention.o_proj.weight": "model-00001-of-00006.safetensors",
396
+ "vision_tower.transformer.layers.10.attention.q_proj.weight": "model-00001-of-00006.safetensors",
397
+ "vision_tower.transformer.layers.10.attention.v_proj.weight": "model-00001-of-00006.safetensors",
398
+ "vision_tower.transformer.layers.10.attention_norm.weight": "model-00001-of-00006.safetensors",
399
+ "vision_tower.transformer.layers.10.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
400
+ "vision_tower.transformer.layers.10.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
401
+ "vision_tower.transformer.layers.10.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
402
+ "vision_tower.transformer.layers.10.ffn_norm.weight": "model-00001-of-00006.safetensors",
403
+ "vision_tower.transformer.layers.11.attention.k_proj.weight": "model-00001-of-00006.safetensors",
404
+ "vision_tower.transformer.layers.11.attention.o_proj.weight": "model-00001-of-00006.safetensors",
405
+ "vision_tower.transformer.layers.11.attention.q_proj.weight": "model-00001-of-00006.safetensors",
406
+ "vision_tower.transformer.layers.11.attention.v_proj.weight": "model-00001-of-00006.safetensors",
407
+ "vision_tower.transformer.layers.11.attention_norm.weight": "model-00001-of-00006.safetensors",
408
+ "vision_tower.transformer.layers.11.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
409
+ "vision_tower.transformer.layers.11.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
410
+ "vision_tower.transformer.layers.11.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
411
+ "vision_tower.transformer.layers.11.ffn_norm.weight": "model-00001-of-00006.safetensors",
412
+ "vision_tower.transformer.layers.12.attention.k_proj.weight": "model-00001-of-00006.safetensors",
413
+ "vision_tower.transformer.layers.12.attention.o_proj.weight": "model-00001-of-00006.safetensors",
414
+ "vision_tower.transformer.layers.12.attention.q_proj.weight": "model-00001-of-00006.safetensors",
415
+ "vision_tower.transformer.layers.12.attention.v_proj.weight": "model-00001-of-00006.safetensors",
416
+ "vision_tower.transformer.layers.12.attention_norm.weight": "model-00001-of-00006.safetensors",
417
+ "vision_tower.transformer.layers.12.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
418
+ "vision_tower.transformer.layers.12.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
419
+ "vision_tower.transformer.layers.12.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
420
+ "vision_tower.transformer.layers.12.ffn_norm.weight": "model-00001-of-00006.safetensors",
421
+ "vision_tower.transformer.layers.13.attention.k_proj.weight": "model-00001-of-00006.safetensors",
422
+ "vision_tower.transformer.layers.13.attention.o_proj.weight": "model-00001-of-00006.safetensors",
423
+ "vision_tower.transformer.layers.13.attention.q_proj.weight": "model-00001-of-00006.safetensors",
424
+ "vision_tower.transformer.layers.13.attention.v_proj.weight": "model-00001-of-00006.safetensors",
425
+ "vision_tower.transformer.layers.13.attention_norm.weight": "model-00001-of-00006.safetensors",
426
+ "vision_tower.transformer.layers.13.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
427
+ "vision_tower.transformer.layers.13.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
428
+ "vision_tower.transformer.layers.13.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
429
+ "vision_tower.transformer.layers.13.ffn_norm.weight": "model-00001-of-00006.safetensors",
430
+ "vision_tower.transformer.layers.14.attention.k_proj.weight": "model-00001-of-00006.safetensors",
431
+ "vision_tower.transformer.layers.14.attention.o_proj.weight": "model-00001-of-00006.safetensors",
432
+ "vision_tower.transformer.layers.14.attention.q_proj.weight": "model-00001-of-00006.safetensors",
433
+ "vision_tower.transformer.layers.14.attention.v_proj.weight": "model-00001-of-00006.safetensors",
434
+ "vision_tower.transformer.layers.14.attention_norm.weight": "model-00001-of-00006.safetensors",
435
+ "vision_tower.transformer.layers.14.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
436
+ "vision_tower.transformer.layers.14.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
437
+ "vision_tower.transformer.layers.14.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
438
+ "vision_tower.transformer.layers.14.ffn_norm.weight": "model-00001-of-00006.safetensors",
439
+ "vision_tower.transformer.layers.15.attention.k_proj.weight": "model-00001-of-00006.safetensors",
440
+ "vision_tower.transformer.layers.15.attention.o_proj.weight": "model-00001-of-00006.safetensors",
441
+ "vision_tower.transformer.layers.15.attention.q_proj.weight": "model-00001-of-00006.safetensors",
442
+ "vision_tower.transformer.layers.15.attention.v_proj.weight": "model-00001-of-00006.safetensors",
443
+ "vision_tower.transformer.layers.15.attention_norm.weight": "model-00001-of-00006.safetensors",
444
+ "vision_tower.transformer.layers.15.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
445
+ "vision_tower.transformer.layers.15.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
446
+ "vision_tower.transformer.layers.15.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
447
+ "vision_tower.transformer.layers.15.ffn_norm.weight": "model-00001-of-00006.safetensors",
448
+ "vision_tower.transformer.layers.16.attention.k_proj.weight": "model-00001-of-00006.safetensors",
449
+ "vision_tower.transformer.layers.16.attention.o_proj.weight": "model-00001-of-00006.safetensors",
450
+ "vision_tower.transformer.layers.16.attention.q_proj.weight": "model-00001-of-00006.safetensors",
451
+ "vision_tower.transformer.layers.16.attention.v_proj.weight": "model-00001-of-00006.safetensors",
452
+ "vision_tower.transformer.layers.16.attention_norm.weight": "model-00001-of-00006.safetensors",
453
+ "vision_tower.transformer.layers.16.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
454
+ "vision_tower.transformer.layers.16.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
455
+ "vision_tower.transformer.layers.16.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
456
+ "vision_tower.transformer.layers.16.ffn_norm.weight": "model-00001-of-00006.safetensors",
457
+ "vision_tower.transformer.layers.17.attention.k_proj.weight": "model-00001-of-00006.safetensors",
458
+ "vision_tower.transformer.layers.17.attention.o_proj.weight": "model-00001-of-00006.safetensors",
459
+ "vision_tower.transformer.layers.17.attention.q_proj.weight": "model-00001-of-00006.safetensors",
460
+ "vision_tower.transformer.layers.17.attention.v_proj.weight": "model-00001-of-00006.safetensors",
461
+ "vision_tower.transformer.layers.17.attention_norm.weight": "model-00001-of-00006.safetensors",
462
+ "vision_tower.transformer.layers.17.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
463
+ "vision_tower.transformer.layers.17.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
464
+ "vision_tower.transformer.layers.17.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
465
+ "vision_tower.transformer.layers.17.ffn_norm.weight": "model-00001-of-00006.safetensors",
466
+ "vision_tower.transformer.layers.18.attention.k_proj.weight": "model-00001-of-00006.safetensors",
467
+ "vision_tower.transformer.layers.18.attention.o_proj.weight": "model-00001-of-00006.safetensors",
468
+ "vision_tower.transformer.layers.18.attention.q_proj.weight": "model-00001-of-00006.safetensors",
469
+ "vision_tower.transformer.layers.18.attention.v_proj.weight": "model-00001-of-00006.safetensors",
470
+ "vision_tower.transformer.layers.18.attention_norm.weight": "model-00001-of-00006.safetensors",
471
+ "vision_tower.transformer.layers.18.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
472
+ "vision_tower.transformer.layers.18.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
473
+ "vision_tower.transformer.layers.18.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
474
+ "vision_tower.transformer.layers.18.ffn_norm.weight": "model-00001-of-00006.safetensors",
475
+ "vision_tower.transformer.layers.19.attention.k_proj.weight": "model-00001-of-00006.safetensors",
476
+ "vision_tower.transformer.layers.19.attention.o_proj.weight": "model-00001-of-00006.safetensors",
477
+ "vision_tower.transformer.layers.19.attention.q_proj.weight": "model-00001-of-00006.safetensors",
478
+ "vision_tower.transformer.layers.19.attention.v_proj.weight": "model-00001-of-00006.safetensors",
479
+ "vision_tower.transformer.layers.19.attention_norm.weight": "model-00001-of-00006.safetensors",
480
+ "vision_tower.transformer.layers.19.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
481
+ "vision_tower.transformer.layers.19.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
482
+ "vision_tower.transformer.layers.19.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
483
+ "vision_tower.transformer.layers.19.ffn_norm.weight": "model-00001-of-00006.safetensors",
484
+ "vision_tower.transformer.layers.2.attention.k_proj.weight": "model-00001-of-00006.safetensors",
485
+ "vision_tower.transformer.layers.2.attention.o_proj.weight": "model-00001-of-00006.safetensors",
486
+ "vision_tower.transformer.layers.2.attention.q_proj.weight": "model-00001-of-00006.safetensors",
487
+ "vision_tower.transformer.layers.2.attention.v_proj.weight": "model-00001-of-00006.safetensors",
488
+ "vision_tower.transformer.layers.2.attention_norm.weight": "model-00001-of-00006.safetensors",
489
+ "vision_tower.transformer.layers.2.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
490
+ "vision_tower.transformer.layers.2.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
491
+ "vision_tower.transformer.layers.2.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
492
+ "vision_tower.transformer.layers.2.ffn_norm.weight": "model-00001-of-00006.safetensors",
493
+ "vision_tower.transformer.layers.20.attention.k_proj.weight": "model-00001-of-00006.safetensors",
494
+ "vision_tower.transformer.layers.20.attention.o_proj.weight": "model-00001-of-00006.safetensors",
495
+ "vision_tower.transformer.layers.20.attention.q_proj.weight": "model-00001-of-00006.safetensors",
496
+ "vision_tower.transformer.layers.20.attention.v_proj.weight": "model-00001-of-00006.safetensors",
497
+ "vision_tower.transformer.layers.20.attention_norm.weight": "model-00001-of-00006.safetensors",
498
+ "vision_tower.transformer.layers.20.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
499
+ "vision_tower.transformer.layers.20.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
500
+ "vision_tower.transformer.layers.20.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
501
+ "vision_tower.transformer.layers.20.ffn_norm.weight": "model-00001-of-00006.safetensors",
502
+ "vision_tower.transformer.layers.21.attention.k_proj.weight": "model-00001-of-00006.safetensors",
503
+ "vision_tower.transformer.layers.21.attention.o_proj.weight": "model-00001-of-00006.safetensors",
504
+ "vision_tower.transformer.layers.21.attention.q_proj.weight": "model-00001-of-00006.safetensors",
505
+ "vision_tower.transformer.layers.21.attention.v_proj.weight": "model-00001-of-00006.safetensors",
506
+ "vision_tower.transformer.layers.21.attention_norm.weight": "model-00001-of-00006.safetensors",
507
+ "vision_tower.transformer.layers.21.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
508
+ "vision_tower.transformer.layers.21.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
509
+ "vision_tower.transformer.layers.21.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
510
+ "vision_tower.transformer.layers.21.ffn_norm.weight": "model-00001-of-00006.safetensors",
511
+ "vision_tower.transformer.layers.22.attention.k_proj.weight": "model-00001-of-00006.safetensors",
512
+ "vision_tower.transformer.layers.22.attention.o_proj.weight": "model-00001-of-00006.safetensors",
513
+ "vision_tower.transformer.layers.22.attention.q_proj.weight": "model-00001-of-00006.safetensors",
514
+ "vision_tower.transformer.layers.22.attention.v_proj.weight": "model-00001-of-00006.safetensors",
515
+ "vision_tower.transformer.layers.22.attention_norm.weight": "model-00001-of-00006.safetensors",
516
+ "vision_tower.transformer.layers.22.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
517
+ "vision_tower.transformer.layers.22.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
518
+ "vision_tower.transformer.layers.22.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
519
+ "vision_tower.transformer.layers.22.ffn_norm.weight": "model-00001-of-00006.safetensors",
520
+ "vision_tower.transformer.layers.23.attention.k_proj.weight": "model-00001-of-00006.safetensors",
521
+ "vision_tower.transformer.layers.23.attention.o_proj.weight": "model-00001-of-00006.safetensors",
522
+ "vision_tower.transformer.layers.23.attention.q_proj.weight": "model-00001-of-00006.safetensors",
523
+ "vision_tower.transformer.layers.23.attention.v_proj.weight": "model-00001-of-00006.safetensors",
524
+ "vision_tower.transformer.layers.23.attention_norm.weight": "model-00001-of-00006.safetensors",
525
+ "vision_tower.transformer.layers.23.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
526
+ "vision_tower.transformer.layers.23.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
527
+ "vision_tower.transformer.layers.23.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
528
+ "vision_tower.transformer.layers.23.ffn_norm.weight": "model-00001-of-00006.safetensors",
529
+ "vision_tower.transformer.layers.3.attention.k_proj.weight": "model-00001-of-00006.safetensors",
530
+ "vision_tower.transformer.layers.3.attention.o_proj.weight": "model-00001-of-00006.safetensors",
531
+ "vision_tower.transformer.layers.3.attention.q_proj.weight": "model-00001-of-00006.safetensors",
532
+ "vision_tower.transformer.layers.3.attention.v_proj.weight": "model-00001-of-00006.safetensors",
533
+ "vision_tower.transformer.layers.3.attention_norm.weight": "model-00001-of-00006.safetensors",
534
+ "vision_tower.transformer.layers.3.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
535
+ "vision_tower.transformer.layers.3.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
536
+ "vision_tower.transformer.layers.3.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
537
+ "vision_tower.transformer.layers.3.ffn_norm.weight": "model-00001-of-00006.safetensors",
538
+ "vision_tower.transformer.layers.4.attention.k_proj.weight": "model-00001-of-00006.safetensors",
539
+ "vision_tower.transformer.layers.4.attention.o_proj.weight": "model-00001-of-00006.safetensors",
540
+ "vision_tower.transformer.layers.4.attention.q_proj.weight": "model-00001-of-00006.safetensors",
541
+ "vision_tower.transformer.layers.4.attention.v_proj.weight": "model-00001-of-00006.safetensors",
542
+ "vision_tower.transformer.layers.4.attention_norm.weight": "model-00001-of-00006.safetensors",
543
+ "vision_tower.transformer.layers.4.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
544
+ "vision_tower.transformer.layers.4.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
545
+ "vision_tower.transformer.layers.4.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
546
+ "vision_tower.transformer.layers.4.ffn_norm.weight": "model-00001-of-00006.safetensors",
547
+ "vision_tower.transformer.layers.5.attention.k_proj.weight": "model-00001-of-00006.safetensors",
548
+ "vision_tower.transformer.layers.5.attention.o_proj.weight": "model-00001-of-00006.safetensors",
549
+ "vision_tower.transformer.layers.5.attention.q_proj.weight": "model-00001-of-00006.safetensors",
550
+ "vision_tower.transformer.layers.5.attention.v_proj.weight": "model-00001-of-00006.safetensors",
551
+ "vision_tower.transformer.layers.5.attention_norm.weight": "model-00001-of-00006.safetensors",
552
+ "vision_tower.transformer.layers.5.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
553
+ "vision_tower.transformer.layers.5.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
554
+ "vision_tower.transformer.layers.5.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
555
+ "vision_tower.transformer.layers.5.ffn_norm.weight": "model-00001-of-00006.safetensors",
556
+ "vision_tower.transformer.layers.6.attention.k_proj.weight": "model-00001-of-00006.safetensors",
557
+ "vision_tower.transformer.layers.6.attention.o_proj.weight": "model-00001-of-00006.safetensors",
558
+ "vision_tower.transformer.layers.6.attention.q_proj.weight": "model-00001-of-00006.safetensors",
559
+ "vision_tower.transformer.layers.6.attention.v_proj.weight": "model-00001-of-00006.safetensors",
560
+ "vision_tower.transformer.layers.6.attention_norm.weight": "model-00001-of-00006.safetensors",
561
+ "vision_tower.transformer.layers.6.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
562
+ "vision_tower.transformer.layers.6.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
563
+ "vision_tower.transformer.layers.6.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
564
+ "vision_tower.transformer.layers.6.ffn_norm.weight": "model-00001-of-00006.safetensors",
565
+ "vision_tower.transformer.layers.7.attention.k_proj.weight": "model-00001-of-00006.safetensors",
566
+ "vision_tower.transformer.layers.7.attention.o_proj.weight": "model-00001-of-00006.safetensors",
567
+ "vision_tower.transformer.layers.7.attention.q_proj.weight": "model-00001-of-00006.safetensors",
568
+ "vision_tower.transformer.layers.7.attention.v_proj.weight": "model-00001-of-00006.safetensors",
569
+ "vision_tower.transformer.layers.7.attention_norm.weight": "model-00001-of-00006.safetensors",
570
+ "vision_tower.transformer.layers.7.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
571
+ "vision_tower.transformer.layers.7.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
572
+ "vision_tower.transformer.layers.7.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
573
+ "vision_tower.transformer.layers.7.ffn_norm.weight": "model-00001-of-00006.safetensors",
574
+ "vision_tower.transformer.layers.8.attention.k_proj.weight": "model-00001-of-00006.safetensors",
575
+ "vision_tower.transformer.layers.8.attention.o_proj.weight": "model-00001-of-00006.safetensors",
576
+ "vision_tower.transformer.layers.8.attention.q_proj.weight": "model-00001-of-00006.safetensors",
577
+ "vision_tower.transformer.layers.8.attention.v_proj.weight": "model-00001-of-00006.safetensors",
578
+ "vision_tower.transformer.layers.8.attention_norm.weight": "model-00001-of-00006.safetensors",
579
+ "vision_tower.transformer.layers.8.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
580
+ "vision_tower.transformer.layers.8.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
581
+ "vision_tower.transformer.layers.8.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
582
+ "vision_tower.transformer.layers.8.ffn_norm.weight": "model-00001-of-00006.safetensors",
583
+ "vision_tower.transformer.layers.9.attention.k_proj.weight": "model-00001-of-00006.safetensors",
584
+ "vision_tower.transformer.layers.9.attention.o_proj.weight": "model-00001-of-00006.safetensors",
585
+ "vision_tower.transformer.layers.9.attention.q_proj.weight": "model-00001-of-00006.safetensors",
586
+ "vision_tower.transformer.layers.9.attention.v_proj.weight": "model-00001-of-00006.safetensors",
587
+ "vision_tower.transformer.layers.9.attention_norm.weight": "model-00001-of-00006.safetensors",
588
+ "vision_tower.transformer.layers.9.feed_forward.down_proj.weight": "model-00001-of-00006.safetensors",
589
+ "vision_tower.transformer.layers.9.feed_forward.gate_proj.weight": "model-00001-of-00006.safetensors",
590
+ "vision_tower.transformer.layers.9.feed_forward.up_proj.weight": "model-00001-of-00006.safetensors",
591
+ "vision_tower.transformer.layers.9.ffn_norm.weight": "model-00001-of-00006.safetensors"
592
+ }
593
+ }
processor_config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_break_token": "[IMG_BREAK]",
3
+ "image_end_token": "[IMG_END]",
4
+ "image_processor": {
5
+ "crop_size": null,
6
+ "data_format": "channels_first",
7
+ "device": null,
8
+ "disable_grouping": null,
9
+ "do_center_crop": null,
10
+ "do_convert_rgb": true,
11
+ "do_normalize": true,
12
+ "do_pad": null,
13
+ "do_rescale": true,
14
+ "do_resize": true,
15
+ "image_mean": [
16
+ 0.48145466,
17
+ 0.4578275,
18
+ 0.40821073
19
+ ],
20
+ "image_processor_type": "PixtralImageProcessorFast",
21
+ "image_seq_length": null,
22
+ "image_std": [
23
+ 0.26862954,
24
+ 0.26130258,
25
+ 0.27577711
26
+ ],
27
+ "input_data_format": null,
28
+ "pad_size": null,
29
+ "patch_size": 14,
30
+ "processor_class": "PixtralProcessor",
31
+ "resample": 3,
32
+ "rescale_factor": 0.00392156862745098,
33
+ "return_tensors": null,
34
+ "size": {
35
+ "longest_edge": 1540
36
+ }
37
+ },
38
+ "image_token": "[IMG]",
39
+ "patch_size": 14,
40
+ "processor_class": "PixtralProcessor",
41
+ "spatial_merge_size": 2
42
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:577575622324b2e099e2648be26bdeb5e5815ffe66d7004e9e3ddbf421db6bf1
3
+ size 17078110
tokenizer_config.json ADDED
@@ -0,0 +1,2023 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "additional_special_tokens": [
6
+ "<unk>",
7
+ "<s>",
8
+ "</s>",
9
+ "[INST]",
10
+ "[/INST]",
11
+ "[AVAILABLE_TOOLS]",
12
+ "[/AVAILABLE_TOOLS]",
13
+ "[TOOL_RESULTS]",
14
+ "[/TOOL_RESULTS]",
15
+ "[TOOL_CALLS]",
16
+ "[IMG]",
17
+ "<pad>",
18
+ "[IMG_BREAK]",
19
+ "[IMG_END]",
20
+ "[PREFIX]",
21
+ "[MIDDLE]",
22
+ "[SUFFIX]",
23
+ "[SYSTEM_PROMPT]",
24
+ "[/SYSTEM_PROMPT]",
25
+ "[TOOL_CONTENT]",
26
+ "<SPECIAL_20>",
27
+ "<SPECIAL_21>",
28
+ "<SPECIAL_22>",
29
+ "<SPECIAL_23>",
30
+ "[AUDIO]",
31
+ "[BEGIN_AUDIO]",
32
+ "<SPECIAL_26>",
33
+ "<SPECIAL_27>",
34
+ "<SPECIAL_28>",
35
+ "<SPECIAL_29>",
36
+ "<SPECIAL_30>",
37
+ "<SPECIAL_31>",
38
+ "[ARGS]",
39
+ "[CALL_ID]",
40
+ "[THINK]",
41
+ "[/THINK]",
42
+ "<SPECIAL_36>",
43
+ "<SPECIAL_37>",
44
+ "<SPECIAL_38>",
45
+ "<SPECIAL_39>",
46
+ "<SPECIAL_40>",
47
+ "<SPECIAL_41>",
48
+ "<SPECIAL_42>",
49
+ "<SPECIAL_43>",
50
+ "<SPECIAL_44>",
51
+ "<SPECIAL_45>",
52
+ "<SPECIAL_46>",
53
+ "<SPECIAL_47>",
54
+ "<SPECIAL_48>",
55
+ "<SPECIAL_49>",
56
+ "<SPECIAL_50>",
57
+ "<SPECIAL_51>",
58
+ "<SPECIAL_52>",
59
+ "<SPECIAL_53>",
60
+ "<SPECIAL_54>",
61
+ "<SPECIAL_55>",
62
+ "<SPECIAL_56>",
63
+ "<SPECIAL_57>",
64
+ "<SPECIAL_58>",
65
+ "<SPECIAL_59>",
66
+ "<SPECIAL_60>",
67
+ "<SPECIAL_61>",
68
+ "<SPECIAL_62>",
69
+ "<SPECIAL_63>",
70
+ "<SPECIAL_64>",
71
+ "<SPECIAL_65>",
72
+ "<SPECIAL_66>",
73
+ "<SPECIAL_67>",
74
+ "<SPECIAL_68>",
75
+ "<SPECIAL_69>",
76
+ "<SPECIAL_70>",
77
+ "<SPECIAL_71>",
78
+ "<SPECIAL_72>",
79
+ "<SPECIAL_73>",
80
+ "<SPECIAL_74>",
81
+ "<SPECIAL_75>",
82
+ "<SPECIAL_76>",
83
+ "<SPECIAL_77>",
84
+ "<SPECIAL_78>",
85
+ "<SPECIAL_79>",
86
+ "<SPECIAL_80>",
87
+ "<SPECIAL_81>",
88
+ "<SPECIAL_82>",
89
+ "<SPECIAL_83>",
90
+ "<SPECIAL_84>",
91
+ "<SPECIAL_85>",
92
+ "<SPECIAL_86>",
93
+ "<SPECIAL_87>",
94
+ "<SPECIAL_88>",
95
+ "<SPECIAL_89>",
96
+ "<SPECIAL_90>",
97
+ "<SPECIAL_91>",
98
+ "<SPECIAL_92>",
99
+ "<SPECIAL_93>",
100
+ "<SPECIAL_94>",
101
+ "<SPECIAL_95>",
102
+ "<SPECIAL_96>",
103
+ "<SPECIAL_97>",
104
+ "<SPECIAL_98>",
105
+ "<SPECIAL_99>",
106
+ "<SPECIAL_100>",
107
+ "<SPECIAL_101>",
108
+ "<SPECIAL_102>",
109
+ "<SPECIAL_103>",
110
+ "<SPECIAL_104>",
111
+ "<SPECIAL_105>",
112
+ "<SPECIAL_106>",
113
+ "<SPECIAL_107>",
114
+ "<SPECIAL_108>",
115
+ "<SPECIAL_109>",
116
+ "<SPECIAL_110>",
117
+ "<SPECIAL_111>",
118
+ "<SPECIAL_112>",
119
+ "<SPECIAL_113>",
120
+ "<SPECIAL_114>",
121
+ "<SPECIAL_115>",
122
+ "<SPECIAL_116>",
123
+ "<SPECIAL_117>",
124
+ "<SPECIAL_118>",
125
+ "<SPECIAL_119>",
126
+ "<SPECIAL_120>",
127
+ "<SPECIAL_121>",
128
+ "<SPECIAL_122>",
129
+ "<SPECIAL_123>",
130
+ "<SPECIAL_124>",
131
+ "<SPECIAL_125>",
132
+ "<SPECIAL_126>",
133
+ "<SPECIAL_127>",
134
+ "<SPECIAL_128>",
135
+ "<SPECIAL_129>",
136
+ "<SPECIAL_130>",
137
+ "<SPECIAL_131>",
138
+ "<SPECIAL_132>",
139
+ "<SPECIAL_133>",
140
+ "<SPECIAL_134>",
141
+ "<SPECIAL_135>",
142
+ "<SPECIAL_136>",
143
+ "<SPECIAL_137>",
144
+ "<SPECIAL_138>",
145
+ "<SPECIAL_139>",
146
+ "<SPECIAL_140>",
147
+ "<SPECIAL_141>",
148
+ "<SPECIAL_142>",
149
+ "<SPECIAL_143>",
150
+ "<SPECIAL_144>",
151
+ "<SPECIAL_145>",
152
+ "<SPECIAL_146>",
153
+ "<SPECIAL_147>",
154
+ "<SPECIAL_148>",
155
+ "<SPECIAL_149>",
156
+ "<SPECIAL_150>",
157
+ "<SPECIAL_151>",
158
+ "<SPECIAL_152>",
159
+ "<SPECIAL_153>",
160
+ "<SPECIAL_154>",
161
+ "<SPECIAL_155>",
162
+ "<SPECIAL_156>",
163
+ "<SPECIAL_157>",
164
+ "<SPECIAL_158>",
165
+ "<SPECIAL_159>",
166
+ "<SPECIAL_160>",
167
+ "<SPECIAL_161>",
168
+ "<SPECIAL_162>",
169
+ "<SPECIAL_163>",
170
+ "<SPECIAL_164>",
171
+ "<SPECIAL_165>",
172
+ "<SPECIAL_166>",
173
+ "<SPECIAL_167>",
174
+ "<SPECIAL_168>",
175
+ "<SPECIAL_169>",
176
+ "<SPECIAL_170>",
177
+ "<SPECIAL_171>",
178
+ "<SPECIAL_172>",
179
+ "<SPECIAL_173>",
180
+ "<SPECIAL_174>",
181
+ "<SPECIAL_175>",
182
+ "<SPECIAL_176>",
183
+ "<SPECIAL_177>",
184
+ "<SPECIAL_178>",
185
+ "<SPECIAL_179>",
186
+ "<SPECIAL_180>",
187
+ "<SPECIAL_181>",
188
+ "<SPECIAL_182>",
189
+ "<SPECIAL_183>",
190
+ "<SPECIAL_184>",
191
+ "<SPECIAL_185>",
192
+ "<SPECIAL_186>",
193
+ "<SPECIAL_187>",
194
+ "<SPECIAL_188>",
195
+ "<SPECIAL_189>",
196
+ "<SPECIAL_190>",
197
+ "<SPECIAL_191>",
198
+ "<SPECIAL_192>",
199
+ "<SPECIAL_193>",
200
+ "<SPECIAL_194>",
201
+ "<SPECIAL_195>",
202
+ "<SPECIAL_196>",
203
+ "<SPECIAL_197>",
204
+ "<SPECIAL_198>",
205
+ "<SPECIAL_199>",
206
+ "<SPECIAL_200>",
207
+ "<SPECIAL_201>",
208
+ "<SPECIAL_202>",
209
+ "<SPECIAL_203>",
210
+ "<SPECIAL_204>",
211
+ "<SPECIAL_205>",
212
+ "<SPECIAL_206>",
213
+ "<SPECIAL_207>",
214
+ "<SPECIAL_208>",
215
+ "<SPECIAL_209>",
216
+ "<SPECIAL_210>",
217
+ "<SPECIAL_211>",
218
+ "<SPECIAL_212>",
219
+ "<SPECIAL_213>",
220
+ "<SPECIAL_214>",
221
+ "<SPECIAL_215>",
222
+ "<SPECIAL_216>",
223
+ "<SPECIAL_217>",
224
+ "<SPECIAL_218>",
225
+ "<SPECIAL_219>",
226
+ "<SPECIAL_220>",
227
+ "<SPECIAL_221>",
228
+ "<SPECIAL_222>",
229
+ "<SPECIAL_223>",
230
+ "<SPECIAL_224>",
231
+ "<SPECIAL_225>",
232
+ "<SPECIAL_226>",
233
+ "<SPECIAL_227>",
234
+ "<SPECIAL_228>",
235
+ "<SPECIAL_229>",
236
+ "<SPECIAL_230>",
237
+ "<SPECIAL_231>",
238
+ "<SPECIAL_232>",
239
+ "<SPECIAL_233>",
240
+ "<SPECIAL_234>",
241
+ "<SPECIAL_235>",
242
+ "<SPECIAL_236>",
243
+ "<SPECIAL_237>",
244
+ "<SPECIAL_238>",
245
+ "<SPECIAL_239>",
246
+ "<SPECIAL_240>",
247
+ "<SPECIAL_241>",
248
+ "<SPECIAL_242>",
249
+ "<SPECIAL_243>",
250
+ "<SPECIAL_244>",
251
+ "<SPECIAL_245>",
252
+ "<SPECIAL_246>",
253
+ "<SPECIAL_247>",
254
+ "<SPECIAL_248>",
255
+ "<SPECIAL_249>",
256
+ "<SPECIAL_250>",
257
+ "<SPECIAL_251>",
258
+ "<SPECIAL_252>",
259
+ "<SPECIAL_253>",
260
+ "<SPECIAL_254>",
261
+ "<SPECIAL_255>",
262
+ "<SPECIAL_256>",
263
+ "<SPECIAL_257>",
264
+ "<SPECIAL_258>",
265
+ "<SPECIAL_259>",
266
+ "<SPECIAL_260>",
267
+ "<SPECIAL_261>",
268
+ "<SPECIAL_262>",
269
+ "<SPECIAL_263>",
270
+ "<SPECIAL_264>",
271
+ "<SPECIAL_265>",
272
+ "<SPECIAL_266>",
273
+ "<SPECIAL_267>",
274
+ "<SPECIAL_268>",
275
+ "<SPECIAL_269>",
276
+ "<SPECIAL_270>",
277
+ "<SPECIAL_271>",
278
+ "<SPECIAL_272>",
279
+ "<SPECIAL_273>",
280
+ "<SPECIAL_274>",
281
+ "<SPECIAL_275>",
282
+ "<SPECIAL_276>",
283
+ "<SPECIAL_277>",
284
+ "<SPECIAL_278>",
285
+ "<SPECIAL_279>",
286
+ "<SPECIAL_280>",
287
+ "<SPECIAL_281>",
288
+ "<SPECIAL_282>",
289
+ "<SPECIAL_283>",
290
+ "<SPECIAL_284>",
291
+ "<SPECIAL_285>",
292
+ "<SPECIAL_286>",
293
+ "<SPECIAL_287>",
294
+ "<SPECIAL_288>",
295
+ "<SPECIAL_289>",
296
+ "<SPECIAL_290>",
297
+ "<SPECIAL_291>",
298
+ "<SPECIAL_292>",
299
+ "<SPECIAL_293>",
300
+ "<SPECIAL_294>",
301
+ "<SPECIAL_295>",
302
+ "<SPECIAL_296>",
303
+ "<SPECIAL_297>",
304
+ "<SPECIAL_298>",
305
+ "<SPECIAL_299>",
306
+ "<SPECIAL_300>",
307
+ "<SPECIAL_301>",
308
+ "<SPECIAL_302>",
309
+ "<SPECIAL_303>",
310
+ "<SPECIAL_304>",
311
+ "<SPECIAL_305>",
312
+ "<SPECIAL_306>",
313
+ "<SPECIAL_307>",
314
+ "<SPECIAL_308>",
315
+ "<SPECIAL_309>",
316
+ "<SPECIAL_310>",
317
+ "<SPECIAL_311>",
318
+ "<SPECIAL_312>",
319
+ "<SPECIAL_313>",
320
+ "<SPECIAL_314>",
321
+ "<SPECIAL_315>",
322
+ "<SPECIAL_316>",
323
+ "<SPECIAL_317>",
324
+ "<SPECIAL_318>",
325
+ "<SPECIAL_319>",
326
+ "<SPECIAL_320>",
327
+ "<SPECIAL_321>",
328
+ "<SPECIAL_322>",
329
+ "<SPECIAL_323>",
330
+ "<SPECIAL_324>",
331
+ "<SPECIAL_325>",
332
+ "<SPECIAL_326>",
333
+ "<SPECIAL_327>",
334
+ "<SPECIAL_328>",
335
+ "<SPECIAL_329>",
336
+ "<SPECIAL_330>",
337
+ "<SPECIAL_331>",
338
+ "<SPECIAL_332>",
339
+ "<SPECIAL_333>",
340
+ "<SPECIAL_334>",
341
+ "<SPECIAL_335>",
342
+ "<SPECIAL_336>",
343
+ "<SPECIAL_337>",
344
+ "<SPECIAL_338>",
345
+ "<SPECIAL_339>",
346
+ "<SPECIAL_340>",
347
+ "<SPECIAL_341>",
348
+ "<SPECIAL_342>",
349
+ "<SPECIAL_343>",
350
+ "<SPECIAL_344>",
351
+ "<SPECIAL_345>",
352
+ "<SPECIAL_346>",
353
+ "<SPECIAL_347>",
354
+ "<SPECIAL_348>",
355
+ "<SPECIAL_349>",
356
+ "<SPECIAL_350>",
357
+ "<SPECIAL_351>",
358
+ "<SPECIAL_352>",
359
+ "<SPECIAL_353>",
360
+ "<SPECIAL_354>",
361
+ "<SPECIAL_355>",
362
+ "<SPECIAL_356>",
363
+ "<SPECIAL_357>",
364
+ "<SPECIAL_358>",
365
+ "<SPECIAL_359>",
366
+ "<SPECIAL_360>",
367
+ "<SPECIAL_361>",
368
+ "<SPECIAL_362>",
369
+ "<SPECIAL_363>",
370
+ "<SPECIAL_364>",
371
+ "<SPECIAL_365>",
372
+ "<SPECIAL_366>",
373
+ "<SPECIAL_367>",
374
+ "<SPECIAL_368>",
375
+ "<SPECIAL_369>",
376
+ "<SPECIAL_370>",
377
+ "<SPECIAL_371>",
378
+ "<SPECIAL_372>",
379
+ "<SPECIAL_373>",
380
+ "<SPECIAL_374>",
381
+ "<SPECIAL_375>",
382
+ "<SPECIAL_376>",
383
+ "<SPECIAL_377>",
384
+ "<SPECIAL_378>",
385
+ "<SPECIAL_379>",
386
+ "<SPECIAL_380>",
387
+ "<SPECIAL_381>",
388
+ "<SPECIAL_382>",
389
+ "<SPECIAL_383>",
390
+ "<SPECIAL_384>",
391
+ "<SPECIAL_385>",
392
+ "<SPECIAL_386>",
393
+ "<SPECIAL_387>",
394
+ "<SPECIAL_388>",
395
+ "<SPECIAL_389>",
396
+ "<SPECIAL_390>",
397
+ "<SPECIAL_391>",
398
+ "<SPECIAL_392>",
399
+ "<SPECIAL_393>",
400
+ "<SPECIAL_394>",
401
+ "<SPECIAL_395>",
402
+ "<SPECIAL_396>",
403
+ "<SPECIAL_397>",
404
+ "<SPECIAL_398>",
405
+ "<SPECIAL_399>",
406
+ "<SPECIAL_400>",
407
+ "<SPECIAL_401>",
408
+ "<SPECIAL_402>",
409
+ "<SPECIAL_403>",
410
+ "<SPECIAL_404>",
411
+ "<SPECIAL_405>",
412
+ "<SPECIAL_406>",
413
+ "<SPECIAL_407>",
414
+ "<SPECIAL_408>",
415
+ "<SPECIAL_409>",
416
+ "<SPECIAL_410>",
417
+ "<SPECIAL_411>",
418
+ "<SPECIAL_412>",
419
+ "<SPECIAL_413>",
420
+ "<SPECIAL_414>",
421
+ "<SPECIAL_415>",
422
+ "<SPECIAL_416>",
423
+ "<SPECIAL_417>",
424
+ "<SPECIAL_418>",
425
+ "<SPECIAL_419>",
426
+ "<SPECIAL_420>",
427
+ "<SPECIAL_421>",
428
+ "<SPECIAL_422>",
429
+ "<SPECIAL_423>",
430
+ "<SPECIAL_424>",
431
+ "<SPECIAL_425>",
432
+ "<SPECIAL_426>",
433
+ "<SPECIAL_427>",
434
+ "<SPECIAL_428>",
435
+ "<SPECIAL_429>",
436
+ "<SPECIAL_430>",
437
+ "<SPECIAL_431>",
438
+ "<SPECIAL_432>",
439
+ "<SPECIAL_433>",
440
+ "<SPECIAL_434>",
441
+ "<SPECIAL_435>",
442
+ "<SPECIAL_436>",
443
+ "<SPECIAL_437>",
444
+ "<SPECIAL_438>",
445
+ "<SPECIAL_439>",
446
+ "<SPECIAL_440>",
447
+ "<SPECIAL_441>",
448
+ "<SPECIAL_442>",
449
+ "<SPECIAL_443>",
450
+ "<SPECIAL_444>",
451
+ "<SPECIAL_445>",
452
+ "<SPECIAL_446>",
453
+ "<SPECIAL_447>",
454
+ "<SPECIAL_448>",
455
+ "<SPECIAL_449>",
456
+ "<SPECIAL_450>",
457
+ "<SPECIAL_451>",
458
+ "<SPECIAL_452>",
459
+ "<SPECIAL_453>",
460
+ "<SPECIAL_454>",
461
+ "<SPECIAL_455>",
462
+ "<SPECIAL_456>",
463
+ "<SPECIAL_457>",
464
+ "<SPECIAL_458>",
465
+ "<SPECIAL_459>",
466
+ "<SPECIAL_460>",
467
+ "<SPECIAL_461>",
468
+ "<SPECIAL_462>",
469
+ "<SPECIAL_463>",
470
+ "<SPECIAL_464>",
471
+ "<SPECIAL_465>",
472
+ "<SPECIAL_466>",
473
+ "<SPECIAL_467>",
474
+ "<SPECIAL_468>",
475
+ "<SPECIAL_469>",
476
+ "<SPECIAL_470>",
477
+ "<SPECIAL_471>",
478
+ "<SPECIAL_472>",
479
+ "<SPECIAL_473>",
480
+ "<SPECIAL_474>",
481
+ "<SPECIAL_475>",
482
+ "<SPECIAL_476>",
483
+ "<SPECIAL_477>",
484
+ "<SPECIAL_478>",
485
+ "<SPECIAL_479>",
486
+ "<SPECIAL_480>",
487
+ "<SPECIAL_481>",
488
+ "<SPECIAL_482>",
489
+ "<SPECIAL_483>",
490
+ "<SPECIAL_484>",
491
+ "<SPECIAL_485>",
492
+ "<SPECIAL_486>",
493
+ "<SPECIAL_487>",
494
+ "<SPECIAL_488>",
495
+ "<SPECIAL_489>",
496
+ "<SPECIAL_490>",
497
+ "<SPECIAL_491>",
498
+ "<SPECIAL_492>",
499
+ "<SPECIAL_493>",
500
+ "<SPECIAL_494>",
501
+ "<SPECIAL_495>",
502
+ "<SPECIAL_496>",
503
+ "<SPECIAL_497>",
504
+ "<SPECIAL_498>",
505
+ "<SPECIAL_499>",
506
+ "<SPECIAL_500>",
507
+ "<SPECIAL_501>",
508
+ "<SPECIAL_502>",
509
+ "<SPECIAL_503>",
510
+ "<SPECIAL_504>",
511
+ "<SPECIAL_505>",
512
+ "<SPECIAL_506>",
513
+ "<SPECIAL_507>",
514
+ "<SPECIAL_508>",
515
+ "<SPECIAL_509>",
516
+ "<SPECIAL_510>",
517
+ "<SPECIAL_511>",
518
+ "<SPECIAL_512>",
519
+ "<SPECIAL_513>",
520
+ "<SPECIAL_514>",
521
+ "<SPECIAL_515>",
522
+ "<SPECIAL_516>",
523
+ "<SPECIAL_517>",
524
+ "<SPECIAL_518>",
525
+ "<SPECIAL_519>",
526
+ "<SPECIAL_520>",
527
+ "<SPECIAL_521>",
528
+ "<SPECIAL_522>",
529
+ "<SPECIAL_523>",
530
+ "<SPECIAL_524>",
531
+ "<SPECIAL_525>",
532
+ "<SPECIAL_526>",
533
+ "<SPECIAL_527>",
534
+ "<SPECIAL_528>",
535
+ "<SPECIAL_529>",
536
+ "<SPECIAL_530>",
537
+ "<SPECIAL_531>",
538
+ "<SPECIAL_532>",
539
+ "<SPECIAL_533>",
540
+ "<SPECIAL_534>",
541
+ "<SPECIAL_535>",
542
+ "<SPECIAL_536>",
543
+ "<SPECIAL_537>",
544
+ "<SPECIAL_538>",
545
+ "<SPECIAL_539>",
546
+ "<SPECIAL_540>",
547
+ "<SPECIAL_541>",
548
+ "<SPECIAL_542>",
549
+ "<SPECIAL_543>",
550
+ "<SPECIAL_544>",
551
+ "<SPECIAL_545>",
552
+ "<SPECIAL_546>",
553
+ "<SPECIAL_547>",
554
+ "<SPECIAL_548>",
555
+ "<SPECIAL_549>",
556
+ "<SPECIAL_550>",
557
+ "<SPECIAL_551>",
558
+ "<SPECIAL_552>",
559
+ "<SPECIAL_553>",
560
+ "<SPECIAL_554>",
561
+ "<SPECIAL_555>",
562
+ "<SPECIAL_556>",
563
+ "<SPECIAL_557>",
564
+ "<SPECIAL_558>",
565
+ "<SPECIAL_559>",
566
+ "<SPECIAL_560>",
567
+ "<SPECIAL_561>",
568
+ "<SPECIAL_562>",
569
+ "<SPECIAL_563>",
570
+ "<SPECIAL_564>",
571
+ "<SPECIAL_565>",
572
+ "<SPECIAL_566>",
573
+ "<SPECIAL_567>",
574
+ "<SPECIAL_568>",
575
+ "<SPECIAL_569>",
576
+ "<SPECIAL_570>",
577
+ "<SPECIAL_571>",
578
+ "<SPECIAL_572>",
579
+ "<SPECIAL_573>",
580
+ "<SPECIAL_574>",
581
+ "<SPECIAL_575>",
582
+ "<SPECIAL_576>",
583
+ "<SPECIAL_577>",
584
+ "<SPECIAL_578>",
585
+ "<SPECIAL_579>",
586
+ "<SPECIAL_580>",
587
+ "<SPECIAL_581>",
588
+ "<SPECIAL_582>",
589
+ "<SPECIAL_583>",
590
+ "<SPECIAL_584>",
591
+ "<SPECIAL_585>",
592
+ "<SPECIAL_586>",
593
+ "<SPECIAL_587>",
594
+ "<SPECIAL_588>",
595
+ "<SPECIAL_589>",
596
+ "<SPECIAL_590>",
597
+ "<SPECIAL_591>",
598
+ "<SPECIAL_592>",
599
+ "<SPECIAL_593>",
600
+ "<SPECIAL_594>",
601
+ "<SPECIAL_595>",
602
+ "<SPECIAL_596>",
603
+ "<SPECIAL_597>",
604
+ "<SPECIAL_598>",
605
+ "<SPECIAL_599>",
606
+ "<SPECIAL_600>",
607
+ "<SPECIAL_601>",
608
+ "<SPECIAL_602>",
609
+ "<SPECIAL_603>",
610
+ "<SPECIAL_604>",
611
+ "<SPECIAL_605>",
612
+ "<SPECIAL_606>",
613
+ "<SPECIAL_607>",
614
+ "<SPECIAL_608>",
615
+ "<SPECIAL_609>",
616
+ "<SPECIAL_610>",
617
+ "<SPECIAL_611>",
618
+ "<SPECIAL_612>",
619
+ "<SPECIAL_613>",
620
+ "<SPECIAL_614>",
621
+ "<SPECIAL_615>",
622
+ "<SPECIAL_616>",
623
+ "<SPECIAL_617>",
624
+ "<SPECIAL_618>",
625
+ "<SPECIAL_619>",
626
+ "<SPECIAL_620>",
627
+ "<SPECIAL_621>",
628
+ "<SPECIAL_622>",
629
+ "<SPECIAL_623>",
630
+ "<SPECIAL_624>",
631
+ "<SPECIAL_625>",
632
+ "<SPECIAL_626>",
633
+ "<SPECIAL_627>",
634
+ "<SPECIAL_628>",
635
+ "<SPECIAL_629>",
636
+ "<SPECIAL_630>",
637
+ "<SPECIAL_631>",
638
+ "<SPECIAL_632>",
639
+ "<SPECIAL_633>",
640
+ "<SPECIAL_634>",
641
+ "<SPECIAL_635>",
642
+ "<SPECIAL_636>",
643
+ "<SPECIAL_637>",
644
+ "<SPECIAL_638>",
645
+ "<SPECIAL_639>",
646
+ "<SPECIAL_640>",
647
+ "<SPECIAL_641>",
648
+ "<SPECIAL_642>",
649
+ "<SPECIAL_643>",
650
+ "<SPECIAL_644>",
651
+ "<SPECIAL_645>",
652
+ "<SPECIAL_646>",
653
+ "<SPECIAL_647>",
654
+ "<SPECIAL_648>",
655
+ "<SPECIAL_649>",
656
+ "<SPECIAL_650>",
657
+ "<SPECIAL_651>",
658
+ "<SPECIAL_652>",
659
+ "<SPECIAL_653>",
660
+ "<SPECIAL_654>",
661
+ "<SPECIAL_655>",
662
+ "<SPECIAL_656>",
663
+ "<SPECIAL_657>",
664
+ "<SPECIAL_658>",
665
+ "<SPECIAL_659>",
666
+ "<SPECIAL_660>",
667
+ "<SPECIAL_661>",
668
+ "<SPECIAL_662>",
669
+ "<SPECIAL_663>",
670
+ "<SPECIAL_664>",
671
+ "<SPECIAL_665>",
672
+ "<SPECIAL_666>",
673
+ "<SPECIAL_667>",
674
+ "<SPECIAL_668>",
675
+ "<SPECIAL_669>",
676
+ "<SPECIAL_670>",
677
+ "<SPECIAL_671>",
678
+ "<SPECIAL_672>",
679
+ "<SPECIAL_673>",
680
+ "<SPECIAL_674>",
681
+ "<SPECIAL_675>",
682
+ "<SPECIAL_676>",
683
+ "<SPECIAL_677>",
684
+ "<SPECIAL_678>",
685
+ "<SPECIAL_679>",
686
+ "<SPECIAL_680>",
687
+ "<SPECIAL_681>",
688
+ "<SPECIAL_682>",
689
+ "<SPECIAL_683>",
690
+ "<SPECIAL_684>",
691
+ "<SPECIAL_685>",
692
+ "<SPECIAL_686>",
693
+ "<SPECIAL_687>",
694
+ "<SPECIAL_688>",
695
+ "<SPECIAL_689>",
696
+ "<SPECIAL_690>",
697
+ "<SPECIAL_691>",
698
+ "<SPECIAL_692>",
699
+ "<SPECIAL_693>",
700
+ "<SPECIAL_694>",
701
+ "<SPECIAL_695>",
702
+ "<SPECIAL_696>",
703
+ "<SPECIAL_697>",
704
+ "<SPECIAL_698>",
705
+ "<SPECIAL_699>",
706
+ "<SPECIAL_700>",
707
+ "<SPECIAL_701>",
708
+ "<SPECIAL_702>",
709
+ "<SPECIAL_703>",
710
+ "<SPECIAL_704>",
711
+ "<SPECIAL_705>",
712
+ "<SPECIAL_706>",
713
+ "<SPECIAL_707>",
714
+ "<SPECIAL_708>",
715
+ "<SPECIAL_709>",
716
+ "<SPECIAL_710>",
717
+ "<SPECIAL_711>",
718
+ "<SPECIAL_712>",
719
+ "<SPECIAL_713>",
720
+ "<SPECIAL_714>",
721
+ "<SPECIAL_715>",
722
+ "<SPECIAL_716>",
723
+ "<SPECIAL_717>",
724
+ "<SPECIAL_718>",
725
+ "<SPECIAL_719>",
726
+ "<SPECIAL_720>",
727
+ "<SPECIAL_721>",
728
+ "<SPECIAL_722>",
729
+ "<SPECIAL_723>",
730
+ "<SPECIAL_724>",
731
+ "<SPECIAL_725>",
732
+ "<SPECIAL_726>",
733
+ "<SPECIAL_727>",
734
+ "<SPECIAL_728>",
735
+ "<SPECIAL_729>",
736
+ "<SPECIAL_730>",
737
+ "<SPECIAL_731>",
738
+ "<SPECIAL_732>",
739
+ "<SPECIAL_733>",
740
+ "<SPECIAL_734>",
741
+ "<SPECIAL_735>",
742
+ "<SPECIAL_736>",
743
+ "<SPECIAL_737>",
744
+ "<SPECIAL_738>",
745
+ "<SPECIAL_739>",
746
+ "<SPECIAL_740>",
747
+ "<SPECIAL_741>",
748
+ "<SPECIAL_742>",
749
+ "<SPECIAL_743>",
750
+ "<SPECIAL_744>",
751
+ "<SPECIAL_745>",
752
+ "<SPECIAL_746>",
753
+ "<SPECIAL_747>",
754
+ "<SPECIAL_748>",
755
+ "<SPECIAL_749>",
756
+ "<SPECIAL_750>",
757
+ "<SPECIAL_751>",
758
+ "<SPECIAL_752>",
759
+ "<SPECIAL_753>",
760
+ "<SPECIAL_754>",
761
+ "<SPECIAL_755>",
762
+ "<SPECIAL_756>",
763
+ "<SPECIAL_757>",
764
+ "<SPECIAL_758>",
765
+ "<SPECIAL_759>",
766
+ "<SPECIAL_760>",
767
+ "<SPECIAL_761>",
768
+ "<SPECIAL_762>",
769
+ "<SPECIAL_763>",
770
+ "<SPECIAL_764>",
771
+ "<SPECIAL_765>",
772
+ "<SPECIAL_766>",
773
+ "<SPECIAL_767>",
774
+ "<SPECIAL_768>",
775
+ "<SPECIAL_769>",
776
+ "<SPECIAL_770>",
777
+ "<SPECIAL_771>",
778
+ "<SPECIAL_772>",
779
+ "<SPECIAL_773>",
780
+ "<SPECIAL_774>",
781
+ "<SPECIAL_775>",
782
+ "<SPECIAL_776>",
783
+ "<SPECIAL_777>",
784
+ "<SPECIAL_778>",
785
+ "<SPECIAL_779>",
786
+ "<SPECIAL_780>",
787
+ "<SPECIAL_781>",
788
+ "<SPECIAL_782>",
789
+ "<SPECIAL_783>",
790
+ "<SPECIAL_784>",
791
+ "<SPECIAL_785>",
792
+ "<SPECIAL_786>",
793
+ "<SPECIAL_787>",
794
+ "<SPECIAL_788>",
795
+ "<SPECIAL_789>",
796
+ "<SPECIAL_790>",
797
+ "<SPECIAL_791>",
798
+ "<SPECIAL_792>",
799
+ "<SPECIAL_793>",
800
+ "<SPECIAL_794>",
801
+ "<SPECIAL_795>",
802
+ "<SPECIAL_796>",
803
+ "<SPECIAL_797>",
804
+ "<SPECIAL_798>",
805
+ "<SPECIAL_799>",
806
+ "<SPECIAL_800>",
807
+ "<SPECIAL_801>",
808
+ "<SPECIAL_802>",
809
+ "<SPECIAL_803>",
810
+ "<SPECIAL_804>",
811
+ "<SPECIAL_805>",
812
+ "<SPECIAL_806>",
813
+ "<SPECIAL_807>",
814
+ "<SPECIAL_808>",
815
+ "<SPECIAL_809>",
816
+ "<SPECIAL_810>",
817
+ "<SPECIAL_811>",
818
+ "<SPECIAL_812>",
819
+ "<SPECIAL_813>",
820
+ "<SPECIAL_814>",
821
+ "<SPECIAL_815>",
822
+ "<SPECIAL_816>",
823
+ "<SPECIAL_817>",
824
+ "<SPECIAL_818>",
825
+ "<SPECIAL_819>",
826
+ "<SPECIAL_820>",
827
+ "<SPECIAL_821>",
828
+ "<SPECIAL_822>",
829
+ "<SPECIAL_823>",
830
+ "<SPECIAL_824>",
831
+ "<SPECIAL_825>",
832
+ "<SPECIAL_826>",
833
+ "<SPECIAL_827>",
834
+ "<SPECIAL_828>",
835
+ "<SPECIAL_829>",
836
+ "<SPECIAL_830>",
837
+ "<SPECIAL_831>",
838
+ "<SPECIAL_832>",
839
+ "<SPECIAL_833>",
840
+ "<SPECIAL_834>",
841
+ "<SPECIAL_835>",
842
+ "<SPECIAL_836>",
843
+ "<SPECIAL_837>",
844
+ "<SPECIAL_838>",
845
+ "<SPECIAL_839>",
846
+ "<SPECIAL_840>",
847
+ "<SPECIAL_841>",
848
+ "<SPECIAL_842>",
849
+ "<SPECIAL_843>",
850
+ "<SPECIAL_844>",
851
+ "<SPECIAL_845>",
852
+ "<SPECIAL_846>",
853
+ "<SPECIAL_847>",
854
+ "<SPECIAL_848>",
855
+ "<SPECIAL_849>",
856
+ "<SPECIAL_850>",
857
+ "<SPECIAL_851>",
858
+ "<SPECIAL_852>",
859
+ "<SPECIAL_853>",
860
+ "<SPECIAL_854>",
861
+ "<SPECIAL_855>",
862
+ "<SPECIAL_856>",
863
+ "<SPECIAL_857>",
864
+ "<SPECIAL_858>",
865
+ "<SPECIAL_859>",
866
+ "<SPECIAL_860>",
867
+ "<SPECIAL_861>",
868
+ "<SPECIAL_862>",
869
+ "<SPECIAL_863>",
870
+ "<SPECIAL_864>",
871
+ "<SPECIAL_865>",
872
+ "<SPECIAL_866>",
873
+ "<SPECIAL_867>",
874
+ "<SPECIAL_868>",
875
+ "<SPECIAL_869>",
876
+ "<SPECIAL_870>",
877
+ "<SPECIAL_871>",
878
+ "<SPECIAL_872>",
879
+ "<SPECIAL_873>",
880
+ "<SPECIAL_874>",
881
+ "<SPECIAL_875>",
882
+ "<SPECIAL_876>",
883
+ "<SPECIAL_877>",
884
+ "<SPECIAL_878>",
885
+ "<SPECIAL_879>",
886
+ "<SPECIAL_880>",
887
+ "<SPECIAL_881>",
888
+ "<SPECIAL_882>",
889
+ "<SPECIAL_883>",
890
+ "<SPECIAL_884>",
891
+ "<SPECIAL_885>",
892
+ "<SPECIAL_886>",
893
+ "<SPECIAL_887>",
894
+ "<SPECIAL_888>",
895
+ "<SPECIAL_889>",
896
+ "<SPECIAL_890>",
897
+ "<SPECIAL_891>",
898
+ "<SPECIAL_892>",
899
+ "<SPECIAL_893>",
900
+ "<SPECIAL_894>",
901
+ "<SPECIAL_895>",
902
+ "<SPECIAL_896>",
903
+ "<SPECIAL_897>",
904
+ "<SPECIAL_898>",
905
+ "<SPECIAL_899>",
906
+ "<SPECIAL_900>",
907
+ "<SPECIAL_901>",
908
+ "<SPECIAL_902>",
909
+ "<SPECIAL_903>",
910
+ "<SPECIAL_904>",
911
+ "<SPECIAL_905>",
912
+ "<SPECIAL_906>",
913
+ "<SPECIAL_907>",
914
+ "<SPECIAL_908>",
915
+ "<SPECIAL_909>",
916
+ "<SPECIAL_910>",
917
+ "<SPECIAL_911>",
918
+ "<SPECIAL_912>",
919
+ "<SPECIAL_913>",
920
+ "<SPECIAL_914>",
921
+ "<SPECIAL_915>",
922
+ "<SPECIAL_916>",
923
+ "<SPECIAL_917>",
924
+ "<SPECIAL_918>",
925
+ "<SPECIAL_919>",
926
+ "<SPECIAL_920>",
927
+ "<SPECIAL_921>",
928
+ "<SPECIAL_922>",
929
+ "<SPECIAL_923>",
930
+ "<SPECIAL_924>",
931
+ "<SPECIAL_925>",
932
+ "<SPECIAL_926>",
933
+ "<SPECIAL_927>",
934
+ "<SPECIAL_928>",
935
+ "<SPECIAL_929>",
936
+ "<SPECIAL_930>",
937
+ "<SPECIAL_931>",
938
+ "<SPECIAL_932>",
939
+ "<SPECIAL_933>",
940
+ "<SPECIAL_934>",
941
+ "<SPECIAL_935>",
942
+ "<SPECIAL_936>",
943
+ "<SPECIAL_937>",
944
+ "<SPECIAL_938>",
945
+ "<SPECIAL_939>",
946
+ "<SPECIAL_940>",
947
+ "<SPECIAL_941>",
948
+ "<SPECIAL_942>",
949
+ "<SPECIAL_943>",
950
+ "<SPECIAL_944>",
951
+ "<SPECIAL_945>",
952
+ "<SPECIAL_946>",
953
+ "<SPECIAL_947>",
954
+ "<SPECIAL_948>",
955
+ "<SPECIAL_949>",
956
+ "<SPECIAL_950>",
957
+ "<SPECIAL_951>",
958
+ "<SPECIAL_952>",
959
+ "<SPECIAL_953>",
960
+ "<SPECIAL_954>",
961
+ "<SPECIAL_955>",
962
+ "<SPECIAL_956>",
963
+ "<SPECIAL_957>",
964
+ "<SPECIAL_958>",
965
+ "<SPECIAL_959>",
966
+ "<SPECIAL_960>",
967
+ "<SPECIAL_961>",
968
+ "<SPECIAL_962>",
969
+ "<SPECIAL_963>",
970
+ "<SPECIAL_964>",
971
+ "<SPECIAL_965>",
972
+ "<SPECIAL_966>",
973
+ "<SPECIAL_967>",
974
+ "<SPECIAL_968>",
975
+ "<SPECIAL_969>",
976
+ "<SPECIAL_970>",
977
+ "<SPECIAL_971>",
978
+ "<SPECIAL_972>",
979
+ "<SPECIAL_973>",
980
+ "<SPECIAL_974>",
981
+ "<SPECIAL_975>",
982
+ "<SPECIAL_976>",
983
+ "<SPECIAL_977>",
984
+ "<SPECIAL_978>",
985
+ "<SPECIAL_979>",
986
+ "<SPECIAL_980>",
987
+ "<SPECIAL_981>",
988
+ "<SPECIAL_982>",
989
+ "<SPECIAL_983>",
990
+ "<SPECIAL_984>",
991
+ "<SPECIAL_985>",
992
+ "<SPECIAL_986>",
993
+ "<SPECIAL_987>",
994
+ "<SPECIAL_988>",
995
+ "<SPECIAL_989>",
996
+ "<SPECIAL_990>",
997
+ "<SPECIAL_991>",
998
+ "<SPECIAL_992>",
999
+ "<SPECIAL_993>",
1000
+ "<SPECIAL_994>",
1001
+ "<SPECIAL_995>",
1002
+ "<SPECIAL_996>",
1003
+ "<SPECIAL_997>",
1004
+ "<SPECIAL_998>",
1005
+ "<SPECIAL_999>"
1006
+ ],
1007
+ "backend": "tokenizers",
1008
+ "bos_token": "<s>",
1009
+ "clean_up_tokenization_spaces": false,
1010
+ "eos_token": "</s>",
1011
+ "extra_special_tokens": [
1012
+ "<unk>",
1013
+ "<s>",
1014
+ "</s>",
1015
+ "[INST]",
1016
+ "[/INST]",
1017
+ "[AVAILABLE_TOOLS]",
1018
+ "[/AVAILABLE_TOOLS]",
1019
+ "[TOOL_RESULTS]",
1020
+ "[/TOOL_RESULTS]",
1021
+ "[TOOL_CALLS]",
1022
+ "[IMG]",
1023
+ "<pad>",
1024
+ "[IMG_BREAK]",
1025
+ "[IMG_END]",
1026
+ "[PREFIX]",
1027
+ "[MIDDLE]",
1028
+ "[SUFFIX]",
1029
+ "[SYSTEM_PROMPT]",
1030
+ "[/SYSTEM_PROMPT]",
1031
+ "[TOOL_CONTENT]",
1032
+ "<SPECIAL_20>",
1033
+ "<SPECIAL_21>",
1034
+ "<SPECIAL_22>",
1035
+ "<SPECIAL_23>",
1036
+ "[AUDIO]",
1037
+ "[BEGIN_AUDIO]",
1038
+ "<SPECIAL_26>",
1039
+ "<SPECIAL_27>",
1040
+ "<SPECIAL_28>",
1041
+ "<SPECIAL_29>",
1042
+ "<SPECIAL_30>",
1043
+ "<SPECIAL_31>",
1044
+ "[ARGS]",
1045
+ "[CALL_ID]",
1046
+ "[THINK]",
1047
+ "[/THINK]",
1048
+ "<SPECIAL_36>",
1049
+ "<SPECIAL_37>",
1050
+ "<SPECIAL_38>",
1051
+ "<SPECIAL_39>",
1052
+ "<SPECIAL_40>",
1053
+ "<SPECIAL_41>",
1054
+ "<SPECIAL_42>",
1055
+ "<SPECIAL_43>",
1056
+ "<SPECIAL_44>",
1057
+ "<SPECIAL_45>",
1058
+ "<SPECIAL_46>",
1059
+ "<SPECIAL_47>",
1060
+ "<SPECIAL_48>",
1061
+ "<SPECIAL_49>",
1062
+ "<SPECIAL_50>",
1063
+ "<SPECIAL_51>",
1064
+ "<SPECIAL_52>",
1065
+ "<SPECIAL_53>",
1066
+ "<SPECIAL_54>",
1067
+ "<SPECIAL_55>",
1068
+ "<SPECIAL_56>",
1069
+ "<SPECIAL_57>",
1070
+ "<SPECIAL_58>",
1071
+ "<SPECIAL_59>",
1072
+ "<SPECIAL_60>",
1073
+ "<SPECIAL_61>",
1074
+ "<SPECIAL_62>",
1075
+ "<SPECIAL_63>",
1076
+ "<SPECIAL_64>",
1077
+ "<SPECIAL_65>",
1078
+ "<SPECIAL_66>",
1079
+ "<SPECIAL_67>",
1080
+ "<SPECIAL_68>",
1081
+ "<SPECIAL_69>",
1082
+ "<SPECIAL_70>",
1083
+ "<SPECIAL_71>",
1084
+ "<SPECIAL_72>",
1085
+ "<SPECIAL_73>",
1086
+ "<SPECIAL_74>",
1087
+ "<SPECIAL_75>",
1088
+ "<SPECIAL_76>",
1089
+ "<SPECIAL_77>",
1090
+ "<SPECIAL_78>",
1091
+ "<SPECIAL_79>",
1092
+ "<SPECIAL_80>",
1093
+ "<SPECIAL_81>",
1094
+ "<SPECIAL_82>",
1095
+ "<SPECIAL_83>",
1096
+ "<SPECIAL_84>",
1097
+ "<SPECIAL_85>",
1098
+ "<SPECIAL_86>",
1099
+ "<SPECIAL_87>",
1100
+ "<SPECIAL_88>",
1101
+ "<SPECIAL_89>",
1102
+ "<SPECIAL_90>",
1103
+ "<SPECIAL_91>",
1104
+ "<SPECIAL_92>",
1105
+ "<SPECIAL_93>",
1106
+ "<SPECIAL_94>",
1107
+ "<SPECIAL_95>",
1108
+ "<SPECIAL_96>",
1109
+ "<SPECIAL_97>",
1110
+ "<SPECIAL_98>",
1111
+ "<SPECIAL_99>",
1112
+ "<SPECIAL_100>",
1113
+ "<SPECIAL_101>",
1114
+ "<SPECIAL_102>",
1115
+ "<SPECIAL_103>",
1116
+ "<SPECIAL_104>",
1117
+ "<SPECIAL_105>",
1118
+ "<SPECIAL_106>",
1119
+ "<SPECIAL_107>",
1120
+ "<SPECIAL_108>",
1121
+ "<SPECIAL_109>",
1122
+ "<SPECIAL_110>",
1123
+ "<SPECIAL_111>",
1124
+ "<SPECIAL_112>",
1125
+ "<SPECIAL_113>",
1126
+ "<SPECIAL_114>",
1127
+ "<SPECIAL_115>",
1128
+ "<SPECIAL_116>",
1129
+ "<SPECIAL_117>",
1130
+ "<SPECIAL_118>",
1131
+ "<SPECIAL_119>",
1132
+ "<SPECIAL_120>",
1133
+ "<SPECIAL_121>",
1134
+ "<SPECIAL_122>",
1135
+ "<SPECIAL_123>",
1136
+ "<SPECIAL_124>",
1137
+ "<SPECIAL_125>",
1138
+ "<SPECIAL_126>",
1139
+ "<SPECIAL_127>",
1140
+ "<SPECIAL_128>",
1141
+ "<SPECIAL_129>",
1142
+ "<SPECIAL_130>",
1143
+ "<SPECIAL_131>",
1144
+ "<SPECIAL_132>",
1145
+ "<SPECIAL_133>",
1146
+ "<SPECIAL_134>",
1147
+ "<SPECIAL_135>",
1148
+ "<SPECIAL_136>",
1149
+ "<SPECIAL_137>",
1150
+ "<SPECIAL_138>",
1151
+ "<SPECIAL_139>",
1152
+ "<SPECIAL_140>",
1153
+ "<SPECIAL_141>",
1154
+ "<SPECIAL_142>",
1155
+ "<SPECIAL_143>",
1156
+ "<SPECIAL_144>",
1157
+ "<SPECIAL_145>",
1158
+ "<SPECIAL_146>",
1159
+ "<SPECIAL_147>",
1160
+ "<SPECIAL_148>",
1161
+ "<SPECIAL_149>",
1162
+ "<SPECIAL_150>",
1163
+ "<SPECIAL_151>",
1164
+ "<SPECIAL_152>",
1165
+ "<SPECIAL_153>",
1166
+ "<SPECIAL_154>",
1167
+ "<SPECIAL_155>",
1168
+ "<SPECIAL_156>",
1169
+ "<SPECIAL_157>",
1170
+ "<SPECIAL_158>",
1171
+ "<SPECIAL_159>",
1172
+ "<SPECIAL_160>",
1173
+ "<SPECIAL_161>",
1174
+ "<SPECIAL_162>",
1175
+ "<SPECIAL_163>",
1176
+ "<SPECIAL_164>",
1177
+ "<SPECIAL_165>",
1178
+ "<SPECIAL_166>",
1179
+ "<SPECIAL_167>",
1180
+ "<SPECIAL_168>",
1181
+ "<SPECIAL_169>",
1182
+ "<SPECIAL_170>",
1183
+ "<SPECIAL_171>",
1184
+ "<SPECIAL_172>",
1185
+ "<SPECIAL_173>",
1186
+ "<SPECIAL_174>",
1187
+ "<SPECIAL_175>",
1188
+ "<SPECIAL_176>",
1189
+ "<SPECIAL_177>",
1190
+ "<SPECIAL_178>",
1191
+ "<SPECIAL_179>",
1192
+ "<SPECIAL_180>",
1193
+ "<SPECIAL_181>",
1194
+ "<SPECIAL_182>",
1195
+ "<SPECIAL_183>",
1196
+ "<SPECIAL_184>",
1197
+ "<SPECIAL_185>",
1198
+ "<SPECIAL_186>",
1199
+ "<SPECIAL_187>",
1200
+ "<SPECIAL_188>",
1201
+ "<SPECIAL_189>",
1202
+ "<SPECIAL_190>",
1203
+ "<SPECIAL_191>",
1204
+ "<SPECIAL_192>",
1205
+ "<SPECIAL_193>",
1206
+ "<SPECIAL_194>",
1207
+ "<SPECIAL_195>",
1208
+ "<SPECIAL_196>",
1209
+ "<SPECIAL_197>",
1210
+ "<SPECIAL_198>",
1211
+ "<SPECIAL_199>",
1212
+ "<SPECIAL_200>",
1213
+ "<SPECIAL_201>",
1214
+ "<SPECIAL_202>",
1215
+ "<SPECIAL_203>",
1216
+ "<SPECIAL_204>",
1217
+ "<SPECIAL_205>",
1218
+ "<SPECIAL_206>",
1219
+ "<SPECIAL_207>",
1220
+ "<SPECIAL_208>",
1221
+ "<SPECIAL_209>",
1222
+ "<SPECIAL_210>",
1223
+ "<SPECIAL_211>",
1224
+ "<SPECIAL_212>",
1225
+ "<SPECIAL_213>",
1226
+ "<SPECIAL_214>",
1227
+ "<SPECIAL_215>",
1228
+ "<SPECIAL_216>",
1229
+ "<SPECIAL_217>",
1230
+ "<SPECIAL_218>",
1231
+ "<SPECIAL_219>",
1232
+ "<SPECIAL_220>",
1233
+ "<SPECIAL_221>",
1234
+ "<SPECIAL_222>",
1235
+ "<SPECIAL_223>",
1236
+ "<SPECIAL_224>",
1237
+ "<SPECIAL_225>",
1238
+ "<SPECIAL_226>",
1239
+ "<SPECIAL_227>",
1240
+ "<SPECIAL_228>",
1241
+ "<SPECIAL_229>",
1242
+ "<SPECIAL_230>",
1243
+ "<SPECIAL_231>",
1244
+ "<SPECIAL_232>",
1245
+ "<SPECIAL_233>",
1246
+ "<SPECIAL_234>",
1247
+ "<SPECIAL_235>",
1248
+ "<SPECIAL_236>",
1249
+ "<SPECIAL_237>",
1250
+ "<SPECIAL_238>",
1251
+ "<SPECIAL_239>",
1252
+ "<SPECIAL_240>",
1253
+ "<SPECIAL_241>",
1254
+ "<SPECIAL_242>",
1255
+ "<SPECIAL_243>",
1256
+ "<SPECIAL_244>",
1257
+ "<SPECIAL_245>",
1258
+ "<SPECIAL_246>",
1259
+ "<SPECIAL_247>",
1260
+ "<SPECIAL_248>",
1261
+ "<SPECIAL_249>",
1262
+ "<SPECIAL_250>",
1263
+ "<SPECIAL_251>",
1264
+ "<SPECIAL_252>",
1265
+ "<SPECIAL_253>",
1266
+ "<SPECIAL_254>",
1267
+ "<SPECIAL_255>",
1268
+ "<SPECIAL_256>",
1269
+ "<SPECIAL_257>",
1270
+ "<SPECIAL_258>",
1271
+ "<SPECIAL_259>",
1272
+ "<SPECIAL_260>",
1273
+ "<SPECIAL_261>",
1274
+ "<SPECIAL_262>",
1275
+ "<SPECIAL_263>",
1276
+ "<SPECIAL_264>",
1277
+ "<SPECIAL_265>",
1278
+ "<SPECIAL_266>",
1279
+ "<SPECIAL_267>",
1280
+ "<SPECIAL_268>",
1281
+ "<SPECIAL_269>",
1282
+ "<SPECIAL_270>",
1283
+ "<SPECIAL_271>",
1284
+ "<SPECIAL_272>",
1285
+ "<SPECIAL_273>",
1286
+ "<SPECIAL_274>",
1287
+ "<SPECIAL_275>",
1288
+ "<SPECIAL_276>",
1289
+ "<SPECIAL_277>",
1290
+ "<SPECIAL_278>",
1291
+ "<SPECIAL_279>",
1292
+ "<SPECIAL_280>",
1293
+ "<SPECIAL_281>",
1294
+ "<SPECIAL_282>",
1295
+ "<SPECIAL_283>",
1296
+ "<SPECIAL_284>",
1297
+ "<SPECIAL_285>",
1298
+ "<SPECIAL_286>",
1299
+ "<SPECIAL_287>",
1300
+ "<SPECIAL_288>",
1301
+ "<SPECIAL_289>",
1302
+ "<SPECIAL_290>",
1303
+ "<SPECIAL_291>",
1304
+ "<SPECIAL_292>",
1305
+ "<SPECIAL_293>",
1306
+ "<SPECIAL_294>",
1307
+ "<SPECIAL_295>",
1308
+ "<SPECIAL_296>",
1309
+ "<SPECIAL_297>",
1310
+ "<SPECIAL_298>",
1311
+ "<SPECIAL_299>",
1312
+ "<SPECIAL_300>",
1313
+ "<SPECIAL_301>",
1314
+ "<SPECIAL_302>",
1315
+ "<SPECIAL_303>",
1316
+ "<SPECIAL_304>",
1317
+ "<SPECIAL_305>",
1318
+ "<SPECIAL_306>",
1319
+ "<SPECIAL_307>",
1320
+ "<SPECIAL_308>",
1321
+ "<SPECIAL_309>",
1322
+ "<SPECIAL_310>",
1323
+ "<SPECIAL_311>",
1324
+ "<SPECIAL_312>",
1325
+ "<SPECIAL_313>",
1326
+ "<SPECIAL_314>",
1327
+ "<SPECIAL_315>",
1328
+ "<SPECIAL_316>",
1329
+ "<SPECIAL_317>",
1330
+ "<SPECIAL_318>",
1331
+ "<SPECIAL_319>",
1332
+ "<SPECIAL_320>",
1333
+ "<SPECIAL_321>",
1334
+ "<SPECIAL_322>",
1335
+ "<SPECIAL_323>",
1336
+ "<SPECIAL_324>",
1337
+ "<SPECIAL_325>",
1338
+ "<SPECIAL_326>",
1339
+ "<SPECIAL_327>",
1340
+ "<SPECIAL_328>",
1341
+ "<SPECIAL_329>",
1342
+ "<SPECIAL_330>",
1343
+ "<SPECIAL_331>",
1344
+ "<SPECIAL_332>",
1345
+ "<SPECIAL_333>",
1346
+ "<SPECIAL_334>",
1347
+ "<SPECIAL_335>",
1348
+ "<SPECIAL_336>",
1349
+ "<SPECIAL_337>",
1350
+ "<SPECIAL_338>",
1351
+ "<SPECIAL_339>",
1352
+ "<SPECIAL_340>",
1353
+ "<SPECIAL_341>",
1354
+ "<SPECIAL_342>",
1355
+ "<SPECIAL_343>",
1356
+ "<SPECIAL_344>",
1357
+ "<SPECIAL_345>",
1358
+ "<SPECIAL_346>",
1359
+ "<SPECIAL_347>",
1360
+ "<SPECIAL_348>",
1361
+ "<SPECIAL_349>",
1362
+ "<SPECIAL_350>",
1363
+ "<SPECIAL_351>",
1364
+ "<SPECIAL_352>",
1365
+ "<SPECIAL_353>",
1366
+ "<SPECIAL_354>",
1367
+ "<SPECIAL_355>",
1368
+ "<SPECIAL_356>",
1369
+ "<SPECIAL_357>",
1370
+ "<SPECIAL_358>",
1371
+ "<SPECIAL_359>",
1372
+ "<SPECIAL_360>",
1373
+ "<SPECIAL_361>",
1374
+ "<SPECIAL_362>",
1375
+ "<SPECIAL_363>",
1376
+ "<SPECIAL_364>",
1377
+ "<SPECIAL_365>",
1378
+ "<SPECIAL_366>",
1379
+ "<SPECIAL_367>",
1380
+ "<SPECIAL_368>",
1381
+ "<SPECIAL_369>",
1382
+ "<SPECIAL_370>",
1383
+ "<SPECIAL_371>",
1384
+ "<SPECIAL_372>",
1385
+ "<SPECIAL_373>",
1386
+ "<SPECIAL_374>",
1387
+ "<SPECIAL_375>",
1388
+ "<SPECIAL_376>",
1389
+ "<SPECIAL_377>",
1390
+ "<SPECIAL_378>",
1391
+ "<SPECIAL_379>",
1392
+ "<SPECIAL_380>",
1393
+ "<SPECIAL_381>",
1394
+ "<SPECIAL_382>",
1395
+ "<SPECIAL_383>",
1396
+ "<SPECIAL_384>",
1397
+ "<SPECIAL_385>",
1398
+ "<SPECIAL_386>",
1399
+ "<SPECIAL_387>",
1400
+ "<SPECIAL_388>",
1401
+ "<SPECIAL_389>",
1402
+ "<SPECIAL_390>",
1403
+ "<SPECIAL_391>",
1404
+ "<SPECIAL_392>",
1405
+ "<SPECIAL_393>",
1406
+ "<SPECIAL_394>",
1407
+ "<SPECIAL_395>",
1408
+ "<SPECIAL_396>",
1409
+ "<SPECIAL_397>",
1410
+ "<SPECIAL_398>",
1411
+ "<SPECIAL_399>",
1412
+ "<SPECIAL_400>",
1413
+ "<SPECIAL_401>",
1414
+ "<SPECIAL_402>",
1415
+ "<SPECIAL_403>",
1416
+ "<SPECIAL_404>",
1417
+ "<SPECIAL_405>",
1418
+ "<SPECIAL_406>",
1419
+ "<SPECIAL_407>",
1420
+ "<SPECIAL_408>",
1421
+ "<SPECIAL_409>",
1422
+ "<SPECIAL_410>",
1423
+ "<SPECIAL_411>",
1424
+ "<SPECIAL_412>",
1425
+ "<SPECIAL_413>",
1426
+ "<SPECIAL_414>",
1427
+ "<SPECIAL_415>",
1428
+ "<SPECIAL_416>",
1429
+ "<SPECIAL_417>",
1430
+ "<SPECIAL_418>",
1431
+ "<SPECIAL_419>",
1432
+ "<SPECIAL_420>",
1433
+ "<SPECIAL_421>",
1434
+ "<SPECIAL_422>",
1435
+ "<SPECIAL_423>",
1436
+ "<SPECIAL_424>",
1437
+ "<SPECIAL_425>",
1438
+ "<SPECIAL_426>",
1439
+ "<SPECIAL_427>",
1440
+ "<SPECIAL_428>",
1441
+ "<SPECIAL_429>",
1442
+ "<SPECIAL_430>",
1443
+ "<SPECIAL_431>",
1444
+ "<SPECIAL_432>",
1445
+ "<SPECIAL_433>",
1446
+ "<SPECIAL_434>",
1447
+ "<SPECIAL_435>",
1448
+ "<SPECIAL_436>",
1449
+ "<SPECIAL_437>",
1450
+ "<SPECIAL_438>",
1451
+ "<SPECIAL_439>",
1452
+ "<SPECIAL_440>",
1453
+ "<SPECIAL_441>",
1454
+ "<SPECIAL_442>",
1455
+ "<SPECIAL_443>",
1456
+ "<SPECIAL_444>",
1457
+ "<SPECIAL_445>",
1458
+ "<SPECIAL_446>",
1459
+ "<SPECIAL_447>",
1460
+ "<SPECIAL_448>",
1461
+ "<SPECIAL_449>",
1462
+ "<SPECIAL_450>",
1463
+ "<SPECIAL_451>",
1464
+ "<SPECIAL_452>",
1465
+ "<SPECIAL_453>",
1466
+ "<SPECIAL_454>",
1467
+ "<SPECIAL_455>",
1468
+ "<SPECIAL_456>",
1469
+ "<SPECIAL_457>",
1470
+ "<SPECIAL_458>",
1471
+ "<SPECIAL_459>",
1472
+ "<SPECIAL_460>",
1473
+ "<SPECIAL_461>",
1474
+ "<SPECIAL_462>",
1475
+ "<SPECIAL_463>",
1476
+ "<SPECIAL_464>",
1477
+ "<SPECIAL_465>",
1478
+ "<SPECIAL_466>",
1479
+ "<SPECIAL_467>",
1480
+ "<SPECIAL_468>",
1481
+ "<SPECIAL_469>",
1482
+ "<SPECIAL_470>",
1483
+ "<SPECIAL_471>",
1484
+ "<SPECIAL_472>",
1485
+ "<SPECIAL_473>",
1486
+ "<SPECIAL_474>",
1487
+ "<SPECIAL_475>",
1488
+ "<SPECIAL_476>",
1489
+ "<SPECIAL_477>",
1490
+ "<SPECIAL_478>",
1491
+ "<SPECIAL_479>",
1492
+ "<SPECIAL_480>",
1493
+ "<SPECIAL_481>",
1494
+ "<SPECIAL_482>",
1495
+ "<SPECIAL_483>",
1496
+ "<SPECIAL_484>",
1497
+ "<SPECIAL_485>",
1498
+ "<SPECIAL_486>",
1499
+ "<SPECIAL_487>",
1500
+ "<SPECIAL_488>",
1501
+ "<SPECIAL_489>",
1502
+ "<SPECIAL_490>",
1503
+ "<SPECIAL_491>",
1504
+ "<SPECIAL_492>",
1505
+ "<SPECIAL_493>",
1506
+ "<SPECIAL_494>",
1507
+ "<SPECIAL_495>",
1508
+ "<SPECIAL_496>",
1509
+ "<SPECIAL_497>",
1510
+ "<SPECIAL_498>",
1511
+ "<SPECIAL_499>",
1512
+ "<SPECIAL_500>",
1513
+ "<SPECIAL_501>",
1514
+ "<SPECIAL_502>",
1515
+ "<SPECIAL_503>",
1516
+ "<SPECIAL_504>",
1517
+ "<SPECIAL_505>",
1518
+ "<SPECIAL_506>",
1519
+ "<SPECIAL_507>",
1520
+ "<SPECIAL_508>",
1521
+ "<SPECIAL_509>",
1522
+ "<SPECIAL_510>",
1523
+ "<SPECIAL_511>",
1524
+ "<SPECIAL_512>",
1525
+ "<SPECIAL_513>",
1526
+ "<SPECIAL_514>",
1527
+ "<SPECIAL_515>",
1528
+ "<SPECIAL_516>",
1529
+ "<SPECIAL_517>",
1530
+ "<SPECIAL_518>",
1531
+ "<SPECIAL_519>",
1532
+ "<SPECIAL_520>",
1533
+ "<SPECIAL_521>",
1534
+ "<SPECIAL_522>",
1535
+ "<SPECIAL_523>",
1536
+ "<SPECIAL_524>",
1537
+ "<SPECIAL_525>",
1538
+ "<SPECIAL_526>",
1539
+ "<SPECIAL_527>",
1540
+ "<SPECIAL_528>",
1541
+ "<SPECIAL_529>",
1542
+ "<SPECIAL_530>",
1543
+ "<SPECIAL_531>",
1544
+ "<SPECIAL_532>",
1545
+ "<SPECIAL_533>",
1546
+ "<SPECIAL_534>",
1547
+ "<SPECIAL_535>",
1548
+ "<SPECIAL_536>",
1549
+ "<SPECIAL_537>",
1550
+ "<SPECIAL_538>",
1551
+ "<SPECIAL_539>",
1552
+ "<SPECIAL_540>",
1553
+ "<SPECIAL_541>",
1554
+ "<SPECIAL_542>",
1555
+ "<SPECIAL_543>",
1556
+ "<SPECIAL_544>",
1557
+ "<SPECIAL_545>",
1558
+ "<SPECIAL_546>",
1559
+ "<SPECIAL_547>",
1560
+ "<SPECIAL_548>",
1561
+ "<SPECIAL_549>",
1562
+ "<SPECIAL_550>",
1563
+ "<SPECIAL_551>",
1564
+ "<SPECIAL_552>",
1565
+ "<SPECIAL_553>",
1566
+ "<SPECIAL_554>",
1567
+ "<SPECIAL_555>",
1568
+ "<SPECIAL_556>",
1569
+ "<SPECIAL_557>",
1570
+ "<SPECIAL_558>",
1571
+ "<SPECIAL_559>",
1572
+ "<SPECIAL_560>",
1573
+ "<SPECIAL_561>",
1574
+ "<SPECIAL_562>",
1575
+ "<SPECIAL_563>",
1576
+ "<SPECIAL_564>",
1577
+ "<SPECIAL_565>",
1578
+ "<SPECIAL_566>",
1579
+ "<SPECIAL_567>",
1580
+ "<SPECIAL_568>",
1581
+ "<SPECIAL_569>",
1582
+ "<SPECIAL_570>",
1583
+ "<SPECIAL_571>",
1584
+ "<SPECIAL_572>",
1585
+ "<SPECIAL_573>",
1586
+ "<SPECIAL_574>",
1587
+ "<SPECIAL_575>",
1588
+ "<SPECIAL_576>",
1589
+ "<SPECIAL_577>",
1590
+ "<SPECIAL_578>",
1591
+ "<SPECIAL_579>",
1592
+ "<SPECIAL_580>",
1593
+ "<SPECIAL_581>",
1594
+ "<SPECIAL_582>",
1595
+ "<SPECIAL_583>",
1596
+ "<SPECIAL_584>",
1597
+ "<SPECIAL_585>",
1598
+ "<SPECIAL_586>",
1599
+ "<SPECIAL_587>",
1600
+ "<SPECIAL_588>",
1601
+ "<SPECIAL_589>",
1602
+ "<SPECIAL_590>",
1603
+ "<SPECIAL_591>",
1604
+ "<SPECIAL_592>",
1605
+ "<SPECIAL_593>",
1606
+ "<SPECIAL_594>",
1607
+ "<SPECIAL_595>",
1608
+ "<SPECIAL_596>",
1609
+ "<SPECIAL_597>",
1610
+ "<SPECIAL_598>",
1611
+ "<SPECIAL_599>",
1612
+ "<SPECIAL_600>",
1613
+ "<SPECIAL_601>",
1614
+ "<SPECIAL_602>",
1615
+ "<SPECIAL_603>",
1616
+ "<SPECIAL_604>",
1617
+ "<SPECIAL_605>",
1618
+ "<SPECIAL_606>",
1619
+ "<SPECIAL_607>",
1620
+ "<SPECIAL_608>",
1621
+ "<SPECIAL_609>",
1622
+ "<SPECIAL_610>",
1623
+ "<SPECIAL_611>",
1624
+ "<SPECIAL_612>",
1625
+ "<SPECIAL_613>",
1626
+ "<SPECIAL_614>",
1627
+ "<SPECIAL_615>",
1628
+ "<SPECIAL_616>",
1629
+ "<SPECIAL_617>",
1630
+ "<SPECIAL_618>",
1631
+ "<SPECIAL_619>",
1632
+ "<SPECIAL_620>",
1633
+ "<SPECIAL_621>",
1634
+ "<SPECIAL_622>",
1635
+ "<SPECIAL_623>",
1636
+ "<SPECIAL_624>",
1637
+ "<SPECIAL_625>",
1638
+ "<SPECIAL_626>",
1639
+ "<SPECIAL_627>",
1640
+ "<SPECIAL_628>",
1641
+ "<SPECIAL_629>",
1642
+ "<SPECIAL_630>",
1643
+ "<SPECIAL_631>",
1644
+ "<SPECIAL_632>",
1645
+ "<SPECIAL_633>",
1646
+ "<SPECIAL_634>",
1647
+ "<SPECIAL_635>",
1648
+ "<SPECIAL_636>",
1649
+ "<SPECIAL_637>",
1650
+ "<SPECIAL_638>",
1651
+ "<SPECIAL_639>",
1652
+ "<SPECIAL_640>",
1653
+ "<SPECIAL_641>",
1654
+ "<SPECIAL_642>",
1655
+ "<SPECIAL_643>",
1656
+ "<SPECIAL_644>",
1657
+ "<SPECIAL_645>",
1658
+ "<SPECIAL_646>",
1659
+ "<SPECIAL_647>",
1660
+ "<SPECIAL_648>",
1661
+ "<SPECIAL_649>",
1662
+ "<SPECIAL_650>",
1663
+ "<SPECIAL_651>",
1664
+ "<SPECIAL_652>",
1665
+ "<SPECIAL_653>",
1666
+ "<SPECIAL_654>",
1667
+ "<SPECIAL_655>",
1668
+ "<SPECIAL_656>",
1669
+ "<SPECIAL_657>",
1670
+ "<SPECIAL_658>",
1671
+ "<SPECIAL_659>",
1672
+ "<SPECIAL_660>",
1673
+ "<SPECIAL_661>",
1674
+ "<SPECIAL_662>",
1675
+ "<SPECIAL_663>",
1676
+ "<SPECIAL_664>",
1677
+ "<SPECIAL_665>",
1678
+ "<SPECIAL_666>",
1679
+ "<SPECIAL_667>",
1680
+ "<SPECIAL_668>",
1681
+ "<SPECIAL_669>",
1682
+ "<SPECIAL_670>",
1683
+ "<SPECIAL_671>",
1684
+ "<SPECIAL_672>",
1685
+ "<SPECIAL_673>",
1686
+ "<SPECIAL_674>",
1687
+ "<SPECIAL_675>",
1688
+ "<SPECIAL_676>",
1689
+ "<SPECIAL_677>",
1690
+ "<SPECIAL_678>",
1691
+ "<SPECIAL_679>",
1692
+ "<SPECIAL_680>",
1693
+ "<SPECIAL_681>",
1694
+ "<SPECIAL_682>",
1695
+ "<SPECIAL_683>",
1696
+ "<SPECIAL_684>",
1697
+ "<SPECIAL_685>",
1698
+ "<SPECIAL_686>",
1699
+ "<SPECIAL_687>",
1700
+ "<SPECIAL_688>",
1701
+ "<SPECIAL_689>",
1702
+ "<SPECIAL_690>",
1703
+ "<SPECIAL_691>",
1704
+ "<SPECIAL_692>",
1705
+ "<SPECIAL_693>",
1706
+ "<SPECIAL_694>",
1707
+ "<SPECIAL_695>",
1708
+ "<SPECIAL_696>",
1709
+ "<SPECIAL_697>",
1710
+ "<SPECIAL_698>",
1711
+ "<SPECIAL_699>",
1712
+ "<SPECIAL_700>",
1713
+ "<SPECIAL_701>",
1714
+ "<SPECIAL_702>",
1715
+ "<SPECIAL_703>",
1716
+ "<SPECIAL_704>",
1717
+ "<SPECIAL_705>",
1718
+ "<SPECIAL_706>",
1719
+ "<SPECIAL_707>",
1720
+ "<SPECIAL_708>",
1721
+ "<SPECIAL_709>",
1722
+ "<SPECIAL_710>",
1723
+ "<SPECIAL_711>",
1724
+ "<SPECIAL_712>",
1725
+ "<SPECIAL_713>",
1726
+ "<SPECIAL_714>",
1727
+ "<SPECIAL_715>",
1728
+ "<SPECIAL_716>",
1729
+ "<SPECIAL_717>",
1730
+ "<SPECIAL_718>",
1731
+ "<SPECIAL_719>",
1732
+ "<SPECIAL_720>",
1733
+ "<SPECIAL_721>",
1734
+ "<SPECIAL_722>",
1735
+ "<SPECIAL_723>",
1736
+ "<SPECIAL_724>",
1737
+ "<SPECIAL_725>",
1738
+ "<SPECIAL_726>",
1739
+ "<SPECIAL_727>",
1740
+ "<SPECIAL_728>",
1741
+ "<SPECIAL_729>",
1742
+ "<SPECIAL_730>",
1743
+ "<SPECIAL_731>",
1744
+ "<SPECIAL_732>",
1745
+ "<SPECIAL_733>",
1746
+ "<SPECIAL_734>",
1747
+ "<SPECIAL_735>",
1748
+ "<SPECIAL_736>",
1749
+ "<SPECIAL_737>",
1750
+ "<SPECIAL_738>",
1751
+ "<SPECIAL_739>",
1752
+ "<SPECIAL_740>",
1753
+ "<SPECIAL_741>",
1754
+ "<SPECIAL_742>",
1755
+ "<SPECIAL_743>",
1756
+ "<SPECIAL_744>",
1757
+ "<SPECIAL_745>",
1758
+ "<SPECIAL_746>",
1759
+ "<SPECIAL_747>",
1760
+ "<SPECIAL_748>",
1761
+ "<SPECIAL_749>",
1762
+ "<SPECIAL_750>",
1763
+ "<SPECIAL_751>",
1764
+ "<SPECIAL_752>",
1765
+ "<SPECIAL_753>",
1766
+ "<SPECIAL_754>",
1767
+ "<SPECIAL_755>",
1768
+ "<SPECIAL_756>",
1769
+ "<SPECIAL_757>",
1770
+ "<SPECIAL_758>",
1771
+ "<SPECIAL_759>",
1772
+ "<SPECIAL_760>",
1773
+ "<SPECIAL_761>",
1774
+ "<SPECIAL_762>",
1775
+ "<SPECIAL_763>",
1776
+ "<SPECIAL_764>",
1777
+ "<SPECIAL_765>",
1778
+ "<SPECIAL_766>",
1779
+ "<SPECIAL_767>",
1780
+ "<SPECIAL_768>",
1781
+ "<SPECIAL_769>",
1782
+ "<SPECIAL_770>",
1783
+ "<SPECIAL_771>",
1784
+ "<SPECIAL_772>",
1785
+ "<SPECIAL_773>",
1786
+ "<SPECIAL_774>",
1787
+ "<SPECIAL_775>",
1788
+ "<SPECIAL_776>",
1789
+ "<SPECIAL_777>",
1790
+ "<SPECIAL_778>",
1791
+ "<SPECIAL_779>",
1792
+ "<SPECIAL_780>",
1793
+ "<SPECIAL_781>",
1794
+ "<SPECIAL_782>",
1795
+ "<SPECIAL_783>",
1796
+ "<SPECIAL_784>",
1797
+ "<SPECIAL_785>",
1798
+ "<SPECIAL_786>",
1799
+ "<SPECIAL_787>",
1800
+ "<SPECIAL_788>",
1801
+ "<SPECIAL_789>",
1802
+ "<SPECIAL_790>",
1803
+ "<SPECIAL_791>",
1804
+ "<SPECIAL_792>",
1805
+ "<SPECIAL_793>",
1806
+ "<SPECIAL_794>",
1807
+ "<SPECIAL_795>",
1808
+ "<SPECIAL_796>",
1809
+ "<SPECIAL_797>",
1810
+ "<SPECIAL_798>",
1811
+ "<SPECIAL_799>",
1812
+ "<SPECIAL_800>",
1813
+ "<SPECIAL_801>",
1814
+ "<SPECIAL_802>",
1815
+ "<SPECIAL_803>",
1816
+ "<SPECIAL_804>",
1817
+ "<SPECIAL_805>",
1818
+ "<SPECIAL_806>",
1819
+ "<SPECIAL_807>",
1820
+ "<SPECIAL_808>",
1821
+ "<SPECIAL_809>",
1822
+ "<SPECIAL_810>",
1823
+ "<SPECIAL_811>",
1824
+ "<SPECIAL_812>",
1825
+ "<SPECIAL_813>",
1826
+ "<SPECIAL_814>",
1827
+ "<SPECIAL_815>",
1828
+ "<SPECIAL_816>",
1829
+ "<SPECIAL_817>",
1830
+ "<SPECIAL_818>",
1831
+ "<SPECIAL_819>",
1832
+ "<SPECIAL_820>",
1833
+ "<SPECIAL_821>",
1834
+ "<SPECIAL_822>",
1835
+ "<SPECIAL_823>",
1836
+ "<SPECIAL_824>",
1837
+ "<SPECIAL_825>",
1838
+ "<SPECIAL_826>",
1839
+ "<SPECIAL_827>",
1840
+ "<SPECIAL_828>",
1841
+ "<SPECIAL_829>",
1842
+ "<SPECIAL_830>",
1843
+ "<SPECIAL_831>",
1844
+ "<SPECIAL_832>",
1845
+ "<SPECIAL_833>",
1846
+ "<SPECIAL_834>",
1847
+ "<SPECIAL_835>",
1848
+ "<SPECIAL_836>",
1849
+ "<SPECIAL_837>",
1850
+ "<SPECIAL_838>",
1851
+ "<SPECIAL_839>",
1852
+ "<SPECIAL_840>",
1853
+ "<SPECIAL_841>",
1854
+ "<SPECIAL_842>",
1855
+ "<SPECIAL_843>",
1856
+ "<SPECIAL_844>",
1857
+ "<SPECIAL_845>",
1858
+ "<SPECIAL_846>",
1859
+ "<SPECIAL_847>",
1860
+ "<SPECIAL_848>",
1861
+ "<SPECIAL_849>",
1862
+ "<SPECIAL_850>",
1863
+ "<SPECIAL_851>",
1864
+ "<SPECIAL_852>",
1865
+ "<SPECIAL_853>",
1866
+ "<SPECIAL_854>",
1867
+ "<SPECIAL_855>",
1868
+ "<SPECIAL_856>",
1869
+ "<SPECIAL_857>",
1870
+ "<SPECIAL_858>",
1871
+ "<SPECIAL_859>",
1872
+ "<SPECIAL_860>",
1873
+ "<SPECIAL_861>",
1874
+ "<SPECIAL_862>",
1875
+ "<SPECIAL_863>",
1876
+ "<SPECIAL_864>",
1877
+ "<SPECIAL_865>",
1878
+ "<SPECIAL_866>",
1879
+ "<SPECIAL_867>",
1880
+ "<SPECIAL_868>",
1881
+ "<SPECIAL_869>",
1882
+ "<SPECIAL_870>",
1883
+ "<SPECIAL_871>",
1884
+ "<SPECIAL_872>",
1885
+ "<SPECIAL_873>",
1886
+ "<SPECIAL_874>",
1887
+ "<SPECIAL_875>",
1888
+ "<SPECIAL_876>",
1889
+ "<SPECIAL_877>",
1890
+ "<SPECIAL_878>",
1891
+ "<SPECIAL_879>",
1892
+ "<SPECIAL_880>",
1893
+ "<SPECIAL_881>",
1894
+ "<SPECIAL_882>",
1895
+ "<SPECIAL_883>",
1896
+ "<SPECIAL_884>",
1897
+ "<SPECIAL_885>",
1898
+ "<SPECIAL_886>",
1899
+ "<SPECIAL_887>",
1900
+ "<SPECIAL_888>",
1901
+ "<SPECIAL_889>",
1902
+ "<SPECIAL_890>",
1903
+ "<SPECIAL_891>",
1904
+ "<SPECIAL_892>",
1905
+ "<SPECIAL_893>",
1906
+ "<SPECIAL_894>",
1907
+ "<SPECIAL_895>",
1908
+ "<SPECIAL_896>",
1909
+ "<SPECIAL_897>",
1910
+ "<SPECIAL_898>",
1911
+ "<SPECIAL_899>",
1912
+ "<SPECIAL_900>",
1913
+ "<SPECIAL_901>",
1914
+ "<SPECIAL_902>",
1915
+ "<SPECIAL_903>",
1916
+ "<SPECIAL_904>",
1917
+ "<SPECIAL_905>",
1918
+ "<SPECIAL_906>",
1919
+ "<SPECIAL_907>",
1920
+ "<SPECIAL_908>",
1921
+ "<SPECIAL_909>",
1922
+ "<SPECIAL_910>",
1923
+ "<SPECIAL_911>",
1924
+ "<SPECIAL_912>",
1925
+ "<SPECIAL_913>",
1926
+ "<SPECIAL_914>",
1927
+ "<SPECIAL_915>",
1928
+ "<SPECIAL_916>",
1929
+ "<SPECIAL_917>",
1930
+ "<SPECIAL_918>",
1931
+ "<SPECIAL_919>",
1932
+ "<SPECIAL_920>",
1933
+ "<SPECIAL_921>",
1934
+ "<SPECIAL_922>",
1935
+ "<SPECIAL_923>",
1936
+ "<SPECIAL_924>",
1937
+ "<SPECIAL_925>",
1938
+ "<SPECIAL_926>",
1939
+ "<SPECIAL_927>",
1940
+ "<SPECIAL_928>",
1941
+ "<SPECIAL_929>",
1942
+ "<SPECIAL_930>",
1943
+ "<SPECIAL_931>",
1944
+ "<SPECIAL_932>",
1945
+ "<SPECIAL_933>",
1946
+ "<SPECIAL_934>",
1947
+ "<SPECIAL_935>",
1948
+ "<SPECIAL_936>",
1949
+ "<SPECIAL_937>",
1950
+ "<SPECIAL_938>",
1951
+ "<SPECIAL_939>",
1952
+ "<SPECIAL_940>",
1953
+ "<SPECIAL_941>",
1954
+ "<SPECIAL_942>",
1955
+ "<SPECIAL_943>",
1956
+ "<SPECIAL_944>",
1957
+ "<SPECIAL_945>",
1958
+ "<SPECIAL_946>",
1959
+ "<SPECIAL_947>",
1960
+ "<SPECIAL_948>",
1961
+ "<SPECIAL_949>",
1962
+ "<SPECIAL_950>",
1963
+ "<SPECIAL_951>",
1964
+ "<SPECIAL_952>",
1965
+ "<SPECIAL_953>",
1966
+ "<SPECIAL_954>",
1967
+ "<SPECIAL_955>",
1968
+ "<SPECIAL_956>",
1969
+ "<SPECIAL_957>",
1970
+ "<SPECIAL_958>",
1971
+ "<SPECIAL_959>",
1972
+ "<SPECIAL_960>",
1973
+ "<SPECIAL_961>",
1974
+ "<SPECIAL_962>",
1975
+ "<SPECIAL_963>",
1976
+ "<SPECIAL_964>",
1977
+ "<SPECIAL_965>",
1978
+ "<SPECIAL_966>",
1979
+ "<SPECIAL_967>",
1980
+ "<SPECIAL_968>",
1981
+ "<SPECIAL_969>",
1982
+ "<SPECIAL_970>",
1983
+ "<SPECIAL_971>",
1984
+ "<SPECIAL_972>",
1985
+ "<SPECIAL_973>",
1986
+ "<SPECIAL_974>",
1987
+ "<SPECIAL_975>",
1988
+ "<SPECIAL_976>",
1989
+ "<SPECIAL_977>",
1990
+ "<SPECIAL_978>",
1991
+ "<SPECIAL_979>",
1992
+ "<SPECIAL_980>",
1993
+ "<SPECIAL_981>",
1994
+ "<SPECIAL_982>",
1995
+ "<SPECIAL_983>",
1996
+ "<SPECIAL_984>",
1997
+ "<SPECIAL_985>",
1998
+ "<SPECIAL_986>",
1999
+ "<SPECIAL_987>",
2000
+ "<SPECIAL_988>",
2001
+ "<SPECIAL_989>",
2002
+ "<SPECIAL_990>",
2003
+ "<SPECIAL_991>",
2004
+ "<SPECIAL_992>",
2005
+ "<SPECIAL_993>",
2006
+ "<SPECIAL_994>",
2007
+ "<SPECIAL_995>",
2008
+ "<SPECIAL_996>",
2009
+ "<SPECIAL_997>",
2010
+ "<SPECIAL_998>",
2011
+ "<SPECIAL_999>"
2012
+ ],
2013
+ "is_local": false,
2014
+ "model_max_length": 262144,
2015
+ "model_specific_special_tokens": {},
2016
+ "pad_token": "<pad>",
2017
+ "padding_side": "left",
2018
+ "processor_class": "PixtralProcessor",
2019
+ "tokenizer_class": "LlamaTokenizer",
2020
+ "unk_token": "<unk>",
2021
+ "use_default_system_prompt": false,
2022
+ "chat_template": "{#- Unsloth template fixes #}\n{#- Default system message if no system prompt is passed. #}\n{%- set default_system_message = '# HOW YOU SHOULD THINK AND ANSWER\\n\\nFirst draft your thinking process (inner monologue) until you arrive at a response. Format your response using Markdown, and use LaTeX for any mathematical equations. Write both your thoughts and the response in the same language as the input.\\n\\nYour thinking process must follow the template below:[THINK]Your thoughts or/and draft, like working through an exercise on scratch paper. Be as casual and as long as you want until you are confident to generate the response to the user.[/THINK]Here, provide a self-contained response.' %}\n\n{#- Begin of sequence token. #}\n{{- bos_token }}\n\n{#- Handle system prompt if it exists. #}\n{#- System prompt supports text content or text and thinking chunks. #}\n{%- if messages[0]['role'] == 'system' %}\n {{- '[SYSTEM_PROMPT]' -}}\n {%- if messages[0]['content'] is string %}\n {{- messages[0]['content'] -}}\n {%- else %} \n {%- for block in messages[0]['content'] %}\n {%- if block['type'] == 'text' %}\n {{- block['text'] }}\n {%- elif block['type'] == 'thinking' %}\n {{- '[THINK]' + block['thinking'] + '[/THINK]' }}\n {%- else %}\n {{- raise_exception('Only text and thinking chunks are supported in system message contents.') }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n {{- '[/SYSTEM_PROMPT]' -}}\n {%- set loop_messages = messages[1:] %}\n{%- else %}\n {%- set loop_messages = messages %}\n {%- if default_system_message != '' %}\n {{- '[SYSTEM_PROMPT]' + default_system_message + '[/SYSTEM_PROMPT]' }}\n {%- endif %}\n{%- endif %}\n\n\n{#- Tools definition #}\n{%- set tools_definition = '' %}\n{%- set has_tools = false %}\n{%- if tools is defined and tools is not none and tools|length > 0 %}\n {%- set has_tools = true %}\n {%- set tools_definition = '[AVAILABLE_TOOLS]' + (tools| tojson) + '[/AVAILABLE_TOOLS]' %}\n {{- tools_definition }}\n{%- endif %}\n\n{#- Checks for alternating user/assistant messages. #}\n{%- set ns = namespace(index=0) %}\n{%- for message in loop_messages %}\n {%- if message.role == 'user' or (message.role == 'assistant' and (message.tool_calls is not defined or message.tool_calls is none or message.tool_calls | length == 0)) %}\n {%- if (message['role'] == 'user') != (ns.index % 2 == 0) %}\n {{- raise_exception('After the optional system message, conversation roles must alternate user and assistant roles except for tool calls and results.') }}\n {%- endif %}\n {%- set ns.index = ns.index + 1 %}\n {%- endif %}\n{%- endfor %}\n\n{#- Handle conversation messages. #}\n{%- for message in loop_messages %}\n\n {#- User messages supports text content or text and image chunks. #}\n {%- if message['role'] == 'user' %}\n {%- if message['content'] is string %}\n {{- '[INST]' + message['content'] + '[/INST]' }}\n {%- elif message['content'] | length > 0 %}\n {{- '[INST]' }}\n {%- if message['content'] | length == 2 %}\n {%- set blocks = message['content'] | sort(attribute='type') %}\n {%- else %}\n {%- set blocks = message['content'] %}\n {%- endif %}\n {%- for block in blocks %}\n {%- if block['type'] == 'text' %}\n {{- block['text'] }}\n {%- elif block['type'] in ['image', 'image_url'] %}\n {{- '[IMG]' }}\n {%- else %}\n {{- raise_exception('Only text, image and image_url chunks are supported in user message content.') }}\n {%- endif %}\n {%- endfor %}\n {{- '[/INST]' }}\n {%- else %}\n {{- raise_exception('User message must have a string or a list of chunks in content') }}\n {%- endif %}\n\n {#- Assistant messages supports text content or text, image and thinking chunks. #}\n {%- elif message['role'] == 'assistant' %}\n\n {%- if message['content'] is string and message['content'] != '' %}\n {{- message['content'] }}\n {%- elif message['content'] is iterable and message['content'] | length > 0 %}\n {%- for block in message['content'] %}\n {%- if block['type'] == 'text' %}\n {{- block['text'] }}\n {%- elif block['type'] == 'thinking' %}\n {{- '[THINK]' + block['thinking'] + '[/THINK]' }}\n {%- else %}\n {{- raise_exception('Only text and thinking chunks are supported in assistant message contents.') }}\n {%- endif %}\n {%- endfor %}\n {%- endif %}\n \n {%- if message['tool_calls'] is defined and message['tool_calls'] is not none and message['tool_calls']|length > 0 %}\n {%- for tool in message['tool_calls'] %}\n {{- '[TOOL_CALLS]' }}\n {%- set name = tool['function']['name'] %}\n {%- set arguments = tool['function']['arguments'] %}\n {%- if arguments is not string %}\n {%- set arguments = arguments|tojson|safe %}\n {%- elif arguments == '' %}\n {%- set arguments = '{}' %}\n {%- endif %}\n {{- name + '[ARGS]' + arguments }}\n {%- endfor %}\n {%- endif %}\n\n {{- eos_token }}\n\n {#- Tool messages only supports text content. #}\n {%- elif message['role'] == 'tool' %}\n {{- '[TOOL_RESULTS]' + message['content']|string + '[/TOOL_RESULTS]' }}\n\n {#- Raise exception for unsupported roles. #}\n {%- else %}\n {{- raise_exception('Only user, assistant and tool roles are supported, got ' + message['role'] + '.') }}\n {%- endif %}\n{%- endfor %}\n\n{#- Copyright 2025-present Unsloth. Apache 2.0 License. #}"
2023
+ }