Commit
·
3966507
verified
·
0
Parent(s):
Super-squash branch 'main' using huggingface_hub
Browse filesCo-authored-by: pandora-s <pandora-s@users.noreply.huggingface.co>
Co-authored-by: juliendenize <juliendenize@users.noreply.huggingface.co>
- .gitattributes +37 -0
- README.md +584 -0
- SYSTEM_PROMPT.txt +5 -0
- chat_template.jinja +126 -0
- config.json +61 -0
- consolidated.safetensors +3 -0
- generation_config.json +7 -0
- model-00001-of-00004.safetensors +3 -0
- model-00002-of-00004.safetensors +3 -0
- model-00003-of-00004.safetensors +3 -0
- model-00004-of-00004.safetensors +3 -0
- model.safetensors.index.json +539 -0
- params.json +47 -0
- processor_config.json +42 -0
- special_tokens_map.json +0 -0
- tekken.json +3 -0
- tokenizer.json +3 -0
- tokenizer_config.json +0 -0
.gitattributes
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
tekken.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,584 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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-3-8B-Base-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 |
+
---
|
| 25 |
+
|
| 26 |
+
# Ministral 3 8B Reasoning 2512
|
| 27 |
+
A balanced model in the Ministral 3 family, **Ministral 3 8B** is a powerful, efficient tiny language model with vision capabilities.
|
| 28 |
+
|
| 29 |
+
This model is the reasoning post-trained version, trained for reasoning tasks, making it ideal for math, coding and stem related use cases.
|
| 30 |
+
|
| 31 |
+
The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware. Ministral 3 8B can even be deployed locally, capable of fitting in 24GB of VRAM in BF16, and less than 12GB of RAM/VRAM when quantized.
|
| 32 |
+
|
| 33 |
+
## Key Features
|
| 34 |
+
Ministral 3 8B consists of two main architectural components:
|
| 35 |
+
- **8.4B Language Model**
|
| 36 |
+
- **0.4B Vision Encoder**
|
| 37 |
+
|
| 38 |
+
The Ministral 3 8B Reasoning model offers the following capabilities:
|
| 39 |
+
- **Vision**: Enables the model to analyze images and provide insights based on visual content, in addition to text.
|
| 40 |
+
- **Multilingual**: Supports dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean, Arabic.
|
| 41 |
+
- **System Prompt**: Maintains strong adherence and support for system prompts.
|
| 42 |
+
- **Agentic**: Offers best-in-class agentic capabilities with native function calling and JSON outputting.
|
| 43 |
+
- **Reasoning**: Excels at complex, multi-step reasoning and dynamic problem-solving.
|
| 44 |
+
- **Edge-Optimized**: Delivers best-in-class performance at a small scale, deployable anywhere.
|
| 45 |
+
- **Apache 2.0 License**: Open-source license allowing usage and modification for both commercial and non-commercial purposes.
|
| 46 |
+
- **Large Context Window**: Supports a 256k context window.
|
| 47 |
+
|
| 48 |
+
### Use Cases
|
| 49 |
+
Perfect for balanced performance in local or embedded systems, combining versatility with efficiency.
|
| 50 |
+
- Chat interfaces in constrained environments
|
| 51 |
+
- Local daily-driver AI assistant
|
| 52 |
+
- Image/document description and understanding
|
| 53 |
+
- Translation and content generation
|
| 54 |
+
- Specialized agentic use cases
|
| 55 |
+
- Fine-tuning and specialization
|
| 56 |
+
- And more...
|
| 57 |
+
|
| 58 |
+
Bringing advanced AI capabilities to resource-constrained environments.
|
| 59 |
+
|
| 60 |
+
## Ministral 3 Family
|
| 61 |
+
|
| 62 |
+
| Model Name | Type | Precision | Link |
|
| 63 |
+
|--------------------------------|--------------------|-----------|------------------------------------------------------------------------------------------|
|
| 64 |
+
| Ministral 3 3B Base 2512 | Base pre-trained | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-3B-Base-2512) |
|
| 65 |
+
| Ministral 3 3B Instruct 2512 | Instruct post-trained | FP8 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-3B-Instruct-2512) |
|
| 66 |
+
| Ministral 3 3B Reasoning 2512 | Reasoning capable | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-3B-Reasoning-2512) |
|
| 67 |
+
| Ministral 3 8B Base 2512 | Base pre-trained | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-8B-Base-2512) |
|
| 68 |
+
| Ministral 3 8B Instruct 2512 | Instruct post-trained | FP8 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-8B-Instruct-2512) |
|
| 69 |
+
| **Ministral 3 8B Reasoning 2512** | **Reasoning capable** | **BF16** | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-8B-Reasoning-2512) |
|
| 70 |
+
| Ministral 3 14B Base 2512 | Base pre-trained | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-14B-Base-2512) |
|
| 71 |
+
| Ministral 3 14B Instruct 2512 | Instruct post-trained | FP8 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512) |
|
| 72 |
+
| Ministral 3 14B Reasoning 2512 | Reasoning capable | BF16 | [Hugging Face](https://huggingface.co/mistralai/Ministral-3-14B-Reasoning-2512) |
|
| 73 |
+
|
| 74 |
+
Other formats available [here](https://huggingface.co/collections/mistralai/ministral-3-additional-checkpoints).
|
| 75 |
+
|
| 76 |
+
## Benchmark Results
|
| 77 |
+
|
| 78 |
+
We compare Ministral 3 to similar sized models.
|
| 79 |
+
|
| 80 |
+
### Reasoning
|
| 81 |
+
|
| 82 |
+
| Model | AIME25 | AIME24 | GPQA Diamond | LiveCodeBench |
|
| 83 |
+
|---------------------------|-------------|-------------|--------------|---------------|
|
| 84 |
+
| **Ministral 3 14B** | <u>0.850</u>| <u>0.898</u>| <u>0.712</u> | <u>0.646</u> |
|
| 85 |
+
| Qwen3-14B (Thinking) | 0.737 | 0.837 | 0.663 | 0.593 |
|
| 86 |
+
| | | | | |
|
| 87 |
+
| **Ministral 3 8B** | 0.787 | <u>0.860</u>| 0.668 | <u>0.616</u> |
|
| 88 |
+
| Qwen3-VL-8B-Thinking | <u>0.798</u>| <u>0.860</u>| <u>0.671</u> | 0.580 |
|
| 89 |
+
| | | | | |
|
| 90 |
+
| **Ministral 3 3B** | <u>0.721</u>| <u>0.775</u>| 0.534 | <u>0.548</u> |
|
| 91 |
+
| Qwen3-VL-4B-Thinking | 0.697 | 0.729 | <u>0.601</u> | 0.513 |
|
| 92 |
+
|
| 93 |
+
### Instruct
|
| 94 |
+
|
| 95 |
+
| Model | Arena Hard | WildBench | MATH Maj@1 | MM MTBench |
|
| 96 |
+
|---------------------------|-------------|------------|-------------|------------------|
|
| 97 |
+
| **Ministral 3 14B** | <u>0.551</u>| <u>68.5</u>| <u>0.904</u>| <u>8.49</u> |
|
| 98 |
+
| Qwen3 14B (Non-Thinking) | 0.427 | 65.1 | 0.870 | NOT MULTIMODAL |
|
| 99 |
+
| Gemma3-12B-Instruct | 0.436 | 63.2 | 0.854 | 6.70 |
|
| 100 |
+
| | | | | |
|
| 101 |
+
| **Ministral 3 8B** | 0.509 | <u>66.8</u>| 0.876 | <u>8.08</u> |
|
| 102 |
+
| Qwen3-VL-8B-Instruct | <u>0.528</u>| 66.3 | <u>0.946</u>| 8.00 |
|
| 103 |
+
| | | | | |
|
| 104 |
+
| **Ministral 3 3B** | 0.305 | <u>56.8</u>| 0.830 | 7.83 |
|
| 105 |
+
| Qwen3-VL-4B-Instruct | <u>0.438</u>| <u>56.8</u>| <u>0.900</u>| <u>8.01</u> |
|
| 106 |
+
| Qwen3-VL-2B-Instruct | 0.163 | 42.2 | 0.786 | 6.36 |
|
| 107 |
+
| Gemma3-4B-Instruct | 0.318 | 49.1 | 0.759 | 5.23 |
|
| 108 |
+
|
| 109 |
+
### Base
|
| 110 |
+
|
| 111 |
+
| Model | Multilingual MMLU | MATH CoT 2-Shot | AGIEval 5-shot | MMLU Redux 5-shot | MMLU 5-shot | TriviaQA 5-shot |
|
| 112 |
+
|---------------------|-------------------|-----------------|----------------|-------------------|-------------|-----------------|
|
| 113 |
+
| **Ministral 3 14B** | 0.742 | <u>0.676</u> | 0.648 | 0.820 | 0.794 | 0.749 |
|
| 114 |
+
| Qwen3 14B Base | <u>0.754</u> | 0.620 | <u>0.661</u> | <u>0.837</u> | <u>0.804</u>| 0.703 |
|
| 115 |
+
| Gemma 3 12B Base | 0.690 | 0.487 | 0.587 | 0.766 | 0.745 | <u>0.788</u> |
|
| 116 |
+
| | | | | | | |
|
| 117 |
+
| **Ministral 3 8B** | <u>0.706</u> | <u>0.626</u> | 0.591 | 0.793 | <u>0.761</u>| <u>0.681</u> |
|
| 118 |
+
| Qwen 3 8B Base | 0.700 | 0.576 | <u>0.596</u> | <u>0.794</u> | 0.760 | 0.639 |
|
| 119 |
+
| | | | | | | |
|
| 120 |
+
| **Ministral 3 3B** | 0.652 | <u>0.601</u> | 0.511 | 0.735 | 0.707 | 0.592 |
|
| 121 |
+
| Qwen 3 4B Base | <u>0.677</u> | 0.405 | <u>0.570</u> | <u>0.759</u> | <u>0.713</u>| 0.530 |
|
| 122 |
+
| Gemma 3 4B Base | 0.516 | 0.294 | 0.430 | 0.626 | 0.589 | <u>0.640</u> |
|
| 123 |
+
|
| 124 |
+
## Usage
|
| 125 |
+
|
| 126 |
+
The model can be used with the following frameworks;
|
| 127 |
+
- [`vllm`](https://github.com/vllm-project/vllm): See [here](#vllm)
|
| 128 |
+
- [`transformers`](https://github.com/huggingface/transformers): See [here](#transformers)
|
| 129 |
+
|
| 130 |
+
### vLLM
|
| 131 |
+
|
| 132 |
+
We recommend using this model with [vLLM](https://github.com/vllm-project/vllm).
|
| 133 |
+
|
| 134 |
+
#### Installation
|
| 135 |
+
|
| 136 |
+
Make sure to install most recent vllm:
|
| 137 |
+
|
| 138 |
+
```
|
| 139 |
+
uv pip install -U vllm \
|
| 140 |
+
--torch-backend=auto \
|
| 141 |
+
--extra-index-url https://wheels.vllm.ai/nightly
|
| 142 |
+
```
|
| 143 |
+
|
| 144 |
+
Doing so should automatically install [`mistral_common >= 1.8.6`](https://github.com/mistralai/mistral-common/releases/tag/v1.8.6).
|
| 145 |
+
|
| 146 |
+
To check:
|
| 147 |
+
```
|
| 148 |
+
python -c "import mistral_common; print(mistral_common.__version__)"
|
| 149 |
+
```
|
| 150 |
+
|
| 151 |
+
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).
|
| 152 |
+
|
| 153 |
+
#### Serve
|
| 154 |
+
|
| 155 |
+
Due to their size, `Ministral-3-3B-Reasoning-2512` and `Ministral-3-8B-Reasoning-2512` can run on a single 1xH200 GPU.
|
| 156 |
+
|
| 157 |
+
A simple launch command is:
|
| 158 |
+
|
| 159 |
+
```bash
|
| 160 |
+
|
| 161 |
+
vllm serve mistralai/Ministral-3-8B-Reasoning-2512 \
|
| 162 |
+
--tokenizer_mode mistral --config_format mistral --load_format mistral \
|
| 163 |
+
--enable-auto-tool-choice --tool-call-parser mistral \
|
| 164 |
+
--reasoning-parser mistral
|
| 165 |
+
```
|
| 166 |
+
|
| 167 |
+
Key parameter notes:
|
| 168 |
+
|
| 169 |
+
* enable-auto-tool-choice: Required when enabling tool usage.
|
| 170 |
+
* tool-call-parser mistral: Required when enabling tool usage.
|
| 171 |
+
* reasoning-parser mistral: Required when enabling reasoning.
|
| 172 |
+
|
| 173 |
+
Additional flags:
|
| 174 |
+
|
| 175 |
+
* 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.
|
| 176 |
+
* You can set `--max-num-batched-tokens` to balance throughput and latency, higher means higher throughput but higher latency.
|
| 177 |
+
|
| 178 |
+
#### Usage of the model
|
| 179 |
+
|
| 180 |
+
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.
|
| 181 |
+
|
| 182 |
+
<details>
|
| 183 |
+
<summary>Vision Reasoning</summary>
|
| 184 |
+
|
| 185 |
+
Let's see if the Ministral 3 model knows when to pick a fight !
|
| 186 |
+
|
| 187 |
+
```python
|
| 188 |
+
from typing import Any
|
| 189 |
+
|
| 190 |
+
from openai import OpenAI
|
| 191 |
+
from huggingface_hub import hf_hub_download
|
| 192 |
+
|
| 193 |
+
# Modify OpenAI's API key and API base to use vLLM's API server.
|
| 194 |
+
openai_api_key = "EMPTY"
|
| 195 |
+
openai_api_base = "http://localhost:8000/v1"
|
| 196 |
+
|
| 197 |
+
TEMP = 0.7
|
| 198 |
+
TOP_P = 0.95
|
| 199 |
+
MAX_TOK = 262144
|
| 200 |
+
client = OpenAI(
|
| 201 |
+
api_key=openai_api_key,
|
| 202 |
+
base_url=openai_api_base,
|
| 203 |
+
)
|
| 204 |
+
|
| 205 |
+
models = client.models.list()
|
| 206 |
+
model = models.data[0].id
|
| 207 |
+
|
| 208 |
+
|
| 209 |
+
def load_system_prompt(repo_id: str, filename: str) -> dict[str, Any]:
|
| 210 |
+
file_path = hf_hub_download(repo_id=repo_id, filename=filename)
|
| 211 |
+
with open(file_path, "r") as file:
|
| 212 |
+
system_prompt = file.read()
|
| 213 |
+
|
| 214 |
+
index_begin_think = system_prompt.find("[THINK]")
|
| 215 |
+
index_end_think = system_prompt.find("[/THINK]")
|
| 216 |
+
|
| 217 |
+
return {
|
| 218 |
+
"role": "system",
|
| 219 |
+
"content": [
|
| 220 |
+
{"type": "text", "text": system_prompt[:index_begin_think]},
|
| 221 |
+
{
|
| 222 |
+
"type": "thinking",
|
| 223 |
+
"thinking": system_prompt[
|
| 224 |
+
index_begin_think + len("[THINK]") : index_end_think
|
| 225 |
+
],
|
| 226 |
+
"closed": True,
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"type": "text",
|
| 230 |
+
"text": system_prompt[index_end_think + len("[/THINK]") :],
|
| 231 |
+
},
|
| 232 |
+
],
|
| 233 |
+
}
|
| 234 |
+
|
| 235 |
+
|
| 236 |
+
SYSTEM_PROMPT = load_system_prompt(model, "SYSTEM_PROMPT.txt")
|
| 237 |
+
|
| 238 |
+
image_url = "https://static.wikia.nocookie.net/essentialsdocs/images/7/70/Battle.png/revision/latest?cb=20220523172438"
|
| 239 |
+
|
| 240 |
+
messages = [
|
| 241 |
+
SYSTEM_PROMPT,
|
| 242 |
+
{
|
| 243 |
+
"role": "user",
|
| 244 |
+
"content": [
|
| 245 |
+
{
|
| 246 |
+
"type": "text",
|
| 247 |
+
"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.",
|
| 248 |
+
},
|
| 249 |
+
{"type": "image_url", "image_url": {"url": image_url}},
|
| 250 |
+
],
|
| 251 |
+
},
|
| 252 |
+
]
|
| 253 |
+
|
| 254 |
+
|
| 255 |
+
stream = client.chat.completions.create(
|
| 256 |
+
model=model,
|
| 257 |
+
messages=messages,
|
| 258 |
+
stream=True,
|
| 259 |
+
temperature=TEMP,
|
| 260 |
+
top_p=TOP_P,
|
| 261 |
+
max_tokens=MAX_TOK,
|
| 262 |
+
)
|
| 263 |
+
|
| 264 |
+
print("client: Start streaming chat completions...:\n")
|
| 265 |
+
printed_reasoning_content = False
|
| 266 |
+
answer = []
|
| 267 |
+
|
| 268 |
+
for chunk in stream:
|
| 269 |
+
reasoning_content = None
|
| 270 |
+
content = None
|
| 271 |
+
# Check the content is reasoning_content or content
|
| 272 |
+
if hasattr(chunk.choices[0].delta, "reasoning_content"):
|
| 273 |
+
reasoning_content = chunk.choices[0].delta.reasoning_content
|
| 274 |
+
if hasattr(chunk.choices[0].delta, "content"):
|
| 275 |
+
content = chunk.choices[0].delta.content
|
| 276 |
+
|
| 277 |
+
if reasoning_content is not None:
|
| 278 |
+
if not printed_reasoning_content:
|
| 279 |
+
printed_reasoning_content = True
|
| 280 |
+
print("Start reasoning:\n", end="", flush=True)
|
| 281 |
+
print(reasoning_content, end="", flush=True)
|
| 282 |
+
elif content is not None:
|
| 283 |
+
# Extract and print the content
|
| 284 |
+
if not reasoning_content and printed_reasoning_content:
|
| 285 |
+
answer.extend(content)
|
| 286 |
+
print(content, end="", flush=True)
|
| 287 |
+
|
| 288 |
+
if answer:
|
| 289 |
+
print("\n\n=============\nAnswer\n=============\n")
|
| 290 |
+
print("".join(answer))
|
| 291 |
+
else:
|
| 292 |
+
print("\n\n=============\nNo Answer\n=============\n")
|
| 293 |
+
print(
|
| 294 |
+
"No answer was generated by the model, probably because the maximum number of tokens was reached."
|
| 295 |
+
)
|
| 296 |
+
```
|
| 297 |
+
|
| 298 |
+
Now we'll make it compute some maths !
|
| 299 |
+
|
| 300 |
+
```python
|
| 301 |
+
from typing import Any
|
| 302 |
+
|
| 303 |
+
from openai import OpenAI
|
| 304 |
+
from huggingface_hub import hf_hub_download
|
| 305 |
+
|
| 306 |
+
# Modify OpenAI's API key and API base to use vLLM's API server.
|
| 307 |
+
openai_api_key = "EMPTY"
|
| 308 |
+
openai_api_base = "http://localhost:8000/v1"
|
| 309 |
+
|
| 310 |
+
TEMP = 0.7
|
| 311 |
+
TOP_P = 0.95
|
| 312 |
+
MAX_TOK = 262144
|
| 313 |
+
client = OpenAI(
|
| 314 |
+
api_key=openai_api_key,
|
| 315 |
+
base_url=openai_api_base,
|
| 316 |
+
)
|
| 317 |
+
|
| 318 |
+
models = client.models.list()
|
| 319 |
+
model = models.data[0].id
|
| 320 |
+
|
| 321 |
+
|
| 322 |
+
def load_system_prompt(repo_id: str, filename: str) -> dict[str, Any]:
|
| 323 |
+
file_path = hf_hub_download(repo_id=repo_id, filename=filename)
|
| 324 |
+
with open(file_path, "r") as file:
|
| 325 |
+
system_prompt = file.read()
|
| 326 |
+
|
| 327 |
+
index_begin_think = system_prompt.find("[THINK]")
|
| 328 |
+
index_end_think = system_prompt.find("[/THINK]")
|
| 329 |
+
|
| 330 |
+
return {
|
| 331 |
+
"role": "system",
|
| 332 |
+
"content": [
|
| 333 |
+
{"type": "text", "text": system_prompt[:index_begin_think]},
|
| 334 |
+
{
|
| 335 |
+
"type": "thinking",
|
| 336 |
+
"thinking": system_prompt[
|
| 337 |
+
index_begin_think + len("[THINK]") : index_end_think
|
| 338 |
+
],
|
| 339 |
+
"closed": True,
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"type": "text",
|
| 343 |
+
"text": system_prompt[index_end_think + len("[/THINK]") :],
|
| 344 |
+
},
|
| 345 |
+
],
|
| 346 |
+
}
|
| 347 |
+
|
| 348 |
+
|
| 349 |
+
SYSTEM_PROMPT = load_system_prompt(model, "SYSTEM_PROMPT.txt")
|
| 350 |
+
|
| 351 |
+
image_url = "https://i.ytimg.com/vi/5Y3xLHeyKZU/hqdefault.jpg"
|
| 352 |
+
|
| 353 |
+
messages = [
|
| 354 |
+
SYSTEM_PROMPT,
|
| 355 |
+
{
|
| 356 |
+
"role": "user",
|
| 357 |
+
"content": [
|
| 358 |
+
{
|
| 359 |
+
"type": "text",
|
| 360 |
+
"text": "Solve the equations. If they contain only numbers, use your calculator, else only think. Answer in the language of the image.",
|
| 361 |
+
},
|
| 362 |
+
{"type": "image_url", "image_url": {"url": image_url}},
|
| 363 |
+
],
|
| 364 |
+
},
|
| 365 |
+
]
|
| 366 |
+
|
| 367 |
+
stream = client.chat.completions.create(
|
| 368 |
+
model=model,
|
| 369 |
+
messages=messages,
|
| 370 |
+
stream=True,
|
| 371 |
+
temperature=TEMP,
|
| 372 |
+
top_p=TOP_P,
|
| 373 |
+
max_tokens=MAX_TOK,
|
| 374 |
+
)
|
| 375 |
+
|
| 376 |
+
print("client: Start streaming chat completions...:\n")
|
| 377 |
+
printed_reasoning_content = False
|
| 378 |
+
answer = []
|
| 379 |
+
|
| 380 |
+
for chunk in stream:
|
| 381 |
+
reasoning_content = None
|
| 382 |
+
content = None
|
| 383 |
+
# Check the content is reasoning_content or content
|
| 384 |
+
if hasattr(chunk.choices[0].delta, "reasoning_content"):
|
| 385 |
+
reasoning_content = chunk.choices[0].delta.reasoning_content
|
| 386 |
+
if hasattr(chunk.choices[0].delta, "content"):
|
| 387 |
+
content = chunk.choices[0].delta.content
|
| 388 |
+
|
| 389 |
+
if reasoning_content is not None:
|
| 390 |
+
if not printed_reasoning_content:
|
| 391 |
+
printed_reasoning_content = True
|
| 392 |
+
print("Start reasoning:\n", end="", flush=True)
|
| 393 |
+
print(reasoning_content, end="", flush=True)
|
| 394 |
+
if content is not None:
|
| 395 |
+
# Extract and print the content
|
| 396 |
+
if not reasoning_content and printed_reasoning_content:
|
| 397 |
+
answer.extend(content)
|
| 398 |
+
print(content, end="", flush=True)
|
| 399 |
+
|
| 400 |
+
if answer:
|
| 401 |
+
print("\n\n=============\nAnswer\n=============\n")
|
| 402 |
+
print("".join(answer))
|
| 403 |
+
else:
|
| 404 |
+
print("\n\n=============\nNo Answer\n=============\n")
|
| 405 |
+
print(
|
| 406 |
+
"No answer was generated by the model, probably because the maximum number of tokens was reached."
|
| 407 |
+
)
|
| 408 |
+
```
|
| 409 |
+
|
| 410 |
+
</details>
|
| 411 |
+
|
| 412 |
+
<details>
|
| 413 |
+
<summary>Text-Only Request</summary>
|
| 414 |
+
|
| 415 |
+
Let's do more maths and leave it up to the model to figure out how to achieve a result.
|
| 416 |
+
|
| 417 |
+
```python
|
| 418 |
+
from typing import Any
|
| 419 |
+
from openai import OpenAI
|
| 420 |
+
from huggingface_hub import hf_hub_download
|
| 421 |
+
|
| 422 |
+
# Modify OpenAI's API key and API base to use vLLM's API server.
|
| 423 |
+
openai_api_key = "EMPTY"
|
| 424 |
+
openai_api_base = "http://localhost:8000/v1"
|
| 425 |
+
|
| 426 |
+
TEMP = 0.7
|
| 427 |
+
TOP_P = 0.95
|
| 428 |
+
MAX_TOK = 262144
|
| 429 |
+
client = OpenAI(
|
| 430 |
+
api_key=openai_api_key,
|
| 431 |
+
base_url=openai_api_base,
|
| 432 |
+
)
|
| 433 |
+
|
| 434 |
+
models = client.models.list()
|
| 435 |
+
model = models.data[0].id
|
| 436 |
+
|
| 437 |
+
|
| 438 |
+
def load_system_prompt(repo_id: str, filename: str) -> dict[str, Any]:
|
| 439 |
+
file_path = hf_hub_download(repo_id=repo_id, filename=filename)
|
| 440 |
+
with open(file_path, "r") as file:
|
| 441 |
+
system_prompt = file.read()
|
| 442 |
+
|
| 443 |
+
index_begin_think = system_prompt.find("[THINK]")
|
| 444 |
+
index_end_think = system_prompt.find("[/THINK]")
|
| 445 |
+
|
| 446 |
+
return {
|
| 447 |
+
"role": "system",
|
| 448 |
+
"content": [
|
| 449 |
+
{"type": "text", "text": system_prompt[:index_begin_think]},
|
| 450 |
+
{
|
| 451 |
+
"type": "thinking",
|
| 452 |
+
"thinking": system_prompt[
|
| 453 |
+
index_begin_think + len("[THINK]") : index_end_think
|
| 454 |
+
],
|
| 455 |
+
"closed": True,
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"type": "text",
|
| 459 |
+
"text": system_prompt[index_end_think + len("[/THINK]") :],
|
| 460 |
+
},
|
| 461 |
+
],
|
| 462 |
+
}
|
| 463 |
+
|
| 464 |
+
|
| 465 |
+
SYSTEM_PROMPT = load_system_prompt(model, "SYSTEM_PROMPT.txt")
|
| 466 |
+
|
| 467 |
+
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."
|
| 468 |
+
|
| 469 |
+
messages = [
|
| 470 |
+
SYSTEM_PROMPT,
|
| 471 |
+
{"role": "user", "content": query}
|
| 472 |
+
]
|
| 473 |
+
stream = client.chat.completions.create(
|
| 474 |
+
model=model,
|
| 475 |
+
messages=messages,
|
| 476 |
+
stream=True,
|
| 477 |
+
temperature=TEMP,
|
| 478 |
+
top_p=TOP_P,
|
| 479 |
+
max_tokens=MAX_TOK,
|
| 480 |
+
)
|
| 481 |
+
|
| 482 |
+
print("client: Start streaming chat completions...:\n")
|
| 483 |
+
printed_reasoning_content = False
|
| 484 |
+
answer = []
|
| 485 |
+
|
| 486 |
+
for chunk in stream:
|
| 487 |
+
reasoning_content = None
|
| 488 |
+
content = None
|
| 489 |
+
# Check the content is reasoning_content or content
|
| 490 |
+
if hasattr(chunk.choices[0].delta, "reasoning_content"):
|
| 491 |
+
reasoning_content = chunk.choices[0].delta.reasoning_content
|
| 492 |
+
if hasattr(chunk.choices[0].delta, "content"):
|
| 493 |
+
content = chunk.choices[0].delta.content
|
| 494 |
+
|
| 495 |
+
if reasoning_content is not None:
|
| 496 |
+
if not printed_reasoning_content:
|
| 497 |
+
printed_reasoning_content = True
|
| 498 |
+
print("Start reasoning:\n", end="", flush=True)
|
| 499 |
+
print(reasoning_content, end="", flush=True)
|
| 500 |
+
if content is not None:
|
| 501 |
+
# Extract and print the content
|
| 502 |
+
if not reasoning_content and printed_reasoning_content:
|
| 503 |
+
answer.extend(content)
|
| 504 |
+
print(content, end="", flush=True)
|
| 505 |
+
|
| 506 |
+
if answer:
|
| 507 |
+
print("\n\n=============\nAnswer\n=============\n")
|
| 508 |
+
print("".join(answer))
|
| 509 |
+
else:
|
| 510 |
+
print("\n\n=============\nNo Answer\n=============\n")
|
| 511 |
+
print("No answer was generated by the model, probably because the maximum number of tokens was reached.")
|
| 512 |
+
```
|
| 513 |
+
|
| 514 |
+
</details>
|
| 515 |
+
|
| 516 |
+
### Transformers
|
| 517 |
+
|
| 518 |
+
You can also use Ministral 3 3B Reasoning 2512 with `Transformers` !
|
| 519 |
+
Make sure to install `Transformers` from its first v5 release candidate or from "main":
|
| 520 |
+
|
| 521 |
+
```
|
| 522 |
+
pip install transformers==5.0.0rc0
|
| 523 |
+
```
|
| 524 |
+
|
| 525 |
+
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.
|
| 526 |
+
|
| 527 |
+
```bash
|
| 528 |
+
pip install mistral-common --upgrade
|
| 529 |
+
```
|
| 530 |
+
|
| 531 |
+
Then load our tokenizer along with the model and generate:
|
| 532 |
+
|
| 533 |
+
<details>
|
| 534 |
+
<summary>Python snippet</summary>
|
| 535 |
+
|
| 536 |
+
```python
|
| 537 |
+
import torch
|
| 538 |
+
from transformers import Mistral3ForConditionalGeneration, MistralCommonBackend
|
| 539 |
+
|
| 540 |
+
model_id = "mistralai/Ministral-3-8B-Reasoning-2512"
|
| 541 |
+
|
| 542 |
+
tokenizer = MistralCommonBackend.from_pretrained(model_id)
|
| 543 |
+
model = Mistral3ForConditionalGeneration.from_pretrained(
|
| 544 |
+
model_id, torch_dtype=torch.bfloat16, device_map="auto"
|
| 545 |
+
)
|
| 546 |
+
|
| 547 |
+
image_url = "https://static.wikia.nocookie.net/essentialsdocs/images/7/70/Battle.png/revision/latest?cb=20220523172438"
|
| 548 |
+
|
| 549 |
+
messages = [
|
| 550 |
+
{
|
| 551 |
+
"role": "user",
|
| 552 |
+
"content": [
|
| 553 |
+
{
|
| 554 |
+
"type": "text",
|
| 555 |
+
"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.",
|
| 556 |
+
},
|
| 557 |
+
{"type": "image_url", "image_url": {"url": image_url}},
|
| 558 |
+
],
|
| 559 |
+
},
|
| 560 |
+
]
|
| 561 |
+
|
| 562 |
+
tokenized = tokenizer.apply_chat_template(messages, return_tensors="pt", return_dict=True)
|
| 563 |
+
|
| 564 |
+
tokenized["input_ids"] = tokenized["input_ids"].to(device="cuda")
|
| 565 |
+
tokenized["pixel_values"] = tokenized["pixel_values"].to(dtype=torch.bfloat16, device="cuda")
|
| 566 |
+
image_sizes = [tokenized["pixel_values"].shape[-2:]]
|
| 567 |
+
|
| 568 |
+
output = model.generate(
|
| 569 |
+
**tokenized,
|
| 570 |
+
image_sizes=image_sizes,
|
| 571 |
+
max_new_tokens=8092,
|
| 572 |
+
)[0]
|
| 573 |
+
|
| 574 |
+
decoded_output = tokenizer.decode(output[len(tokenized["input_ids"][0]):])
|
| 575 |
+
print(decoded_output)
|
| 576 |
+
```
|
| 577 |
+
|
| 578 |
+
</details>
|
| 579 |
+
|
| 580 |
+
## License
|
| 581 |
+
|
| 582 |
+
This model is licensed under the [Apache 2.0 License](https://www.apache.org/licenses/LICENSE-2.0.txt).
|
| 583 |
+
|
| 584 |
+
*You must not use this model in a manner that infringes, misappropriates, or otherwise violates any third party’s rights, including intellectual property rights.*
|
SYSTEM_PROMPT.txt
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# HOW YOU SHOULD THINK AND ANSWER
|
| 2 |
+
|
| 3 |
+
First 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.
|
| 4 |
+
|
| 5 |
+
Your 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.
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,126 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{#- Default system message if no system prompt is passed. #}
|
| 2 |
+
{%- 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.' %}
|
| 3 |
+
|
| 4 |
+
{#- Begin of sequence token. #}
|
| 5 |
+
{{- bos_token }}
|
| 6 |
+
|
| 7 |
+
{#- Handle system prompt if it exists. #}
|
| 8 |
+
{#- System prompt supports text content or text and thinking chunks. #}
|
| 9 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 10 |
+
{{- '[SYSTEM_PROMPT]' -}}
|
| 11 |
+
{%- if messages[0]['content'] is string %}
|
| 12 |
+
{{- messages[0]['content'] -}}
|
| 13 |
+
{%- else %}
|
| 14 |
+
{%- for block in messages[0]['content'] %}
|
| 15 |
+
{%- if block['type'] == 'text' %}
|
| 16 |
+
{{- block['text'] }}
|
| 17 |
+
{%- elif block['type'] == 'thinking' %}
|
| 18 |
+
{{- '[THINK]' + block['thinking'] + '[/THINK]' }}
|
| 19 |
+
{%- else %}
|
| 20 |
+
{{- raise_exception('Only text and thinking chunks are supported in system message contents.') }}
|
| 21 |
+
{%- endif %}
|
| 22 |
+
{%- endfor %}
|
| 23 |
+
{%- endif %}
|
| 24 |
+
{{- '[/SYSTEM_PROMPT]' -}}
|
| 25 |
+
{%- set loop_messages = messages[1:] %}
|
| 26 |
+
{%- else %}
|
| 27 |
+
{%- set loop_messages = messages %}
|
| 28 |
+
{%- if default_system_message != '' %}
|
| 29 |
+
{{- '[SYSTEM_PROMPT]' + default_system_message + '[/SYSTEM_PROMPT]' }}
|
| 30 |
+
{%- endif %}
|
| 31 |
+
{%- endif %}
|
| 32 |
+
|
| 33 |
+
|
| 34 |
+
{#- Tools definition #}
|
| 35 |
+
{%- set tools_definition = '' %}
|
| 36 |
+
{%- set has_tools = false %}
|
| 37 |
+
{%- if tools is defined and tools is not none and tools|length > 0 %}
|
| 38 |
+
{%- set has_tools = true %}
|
| 39 |
+
{%- set tools_definition = '[AVAILABLE_TOOLS]' + (tools| tojson) + '[/AVAILABLE_TOOLS]' %}
|
| 40 |
+
{{- tools_definition }}
|
| 41 |
+
{%- endif %}
|
| 42 |
+
|
| 43 |
+
{#- Checks for alternating user/assistant messages. #}
|
| 44 |
+
{%- set ns = namespace(index=0) %}
|
| 45 |
+
{%- for message in loop_messages %}
|
| 46 |
+
{%- 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)) %}
|
| 47 |
+
{%- if (message['role'] == 'user') != (ns.index % 2 == 0) %}
|
| 48 |
+
{{- raise_exception('After the optional system message, conversation roles must alternate user and assistant roles except for tool calls and results.') }}
|
| 49 |
+
{%- endif %}
|
| 50 |
+
{%- set ns.index = ns.index + 1 %}
|
| 51 |
+
{%- endif %}
|
| 52 |
+
{%- endfor %}
|
| 53 |
+
|
| 54 |
+
{#- Handle conversation messages. #}
|
| 55 |
+
{%- for message in loop_messages %}
|
| 56 |
+
|
| 57 |
+
{#- User messages supports text content or text and image chunks. #}
|
| 58 |
+
{%- if message['role'] == 'user' %}
|
| 59 |
+
{%- if message['content'] is string %}
|
| 60 |
+
{{- '[INST]' + message['content'] + '[/INST]' }}
|
| 61 |
+
{%- elif message['content'] | length > 0 %}
|
| 62 |
+
{{- '[INST]' }}
|
| 63 |
+
{%- if message['content'] | length == 2 %}
|
| 64 |
+
{%- set blocks = message['content'] | sort(attribute='type') %}
|
| 65 |
+
{%- else %}
|
| 66 |
+
{%- set blocks = message['content'] %}
|
| 67 |
+
{%- endif %}
|
| 68 |
+
{%- for block in blocks %}
|
| 69 |
+
{%- if block['type'] == 'text' %}
|
| 70 |
+
{{- block['text'] }}
|
| 71 |
+
{%- elif block['type'] in ['image', 'image_url'] %}
|
| 72 |
+
{{- '[IMG]' }}
|
| 73 |
+
{%- else %}
|
| 74 |
+
{{- raise_exception('Only text, image and image_url chunks are supported in user message content.') }}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endfor %}
|
| 77 |
+
{{- '[/INST]' }}
|
| 78 |
+
{%- else %}
|
| 79 |
+
{{- raise_exception('User message must have a string or a list of chunks in content') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
|
| 82 |
+
{#- Assistant messages supports text content or text, image and thinking chunks. #}
|
| 83 |
+
{%- elif message['role'] == 'assistant' %}
|
| 84 |
+
{%- if (message['content'] is none or message['content'] == '' or message['content']|length == 0) and (message['tool_calls'] is not defined or message['tool_calls'] is none or message['tool_calls']|length == 0) %}
|
| 85 |
+
{{- raise_exception('Assistant message must have a string or a list of chunks in content or a list of tool calls.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
|
| 88 |
+
{%- if message['content'] is string and message['content'] != '' %}
|
| 89 |
+
{{- message['content'] }}
|
| 90 |
+
{%- elif message['content'] | length > 0 %}
|
| 91 |
+
{%- for block in message['content'] %}
|
| 92 |
+
{%- if block['type'] == 'text' %}
|
| 93 |
+
{{- block['text'] }}
|
| 94 |
+
{%- elif block['type'] == 'thinking' %}
|
| 95 |
+
{{- '[THINK]' + block['thinking'] + '[/THINK]' }}
|
| 96 |
+
{%- else %}
|
| 97 |
+
{{- raise_exception('Only text and thinking chunks are supported in assistant message contents.') }}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- endfor %}
|
| 100 |
+
{%- endif %}
|
| 101 |
+
|
| 102 |
+
{%- if message['tool_calls'] is defined and message['tool_calls'] is not none and message['tool_calls']|length > 0 %}
|
| 103 |
+
{%- for tool in message['tool_calls'] %}
|
| 104 |
+
{{- '[TOOL_CALLS]' }}
|
| 105 |
+
{%- set name = tool['function']['name'] %}
|
| 106 |
+
{%- set arguments = tool['function']['arguments'] %}
|
| 107 |
+
{%- if arguments is not string %}
|
| 108 |
+
{%- set arguments = arguments|tojson|safe %}
|
| 109 |
+
{%- elif arguments == '' %}
|
| 110 |
+
{%- set arguments = '{}' %}
|
| 111 |
+
{%- endif %}
|
| 112 |
+
{{- name + '[ARGS]' + arguments }}
|
| 113 |
+
{%- endfor %}
|
| 114 |
+
{%- endif %}
|
| 115 |
+
|
| 116 |
+
{{- eos_token }}
|
| 117 |
+
|
| 118 |
+
{#- Tool messages only supports text content. #}
|
| 119 |
+
{%- elif message['role'] == 'tool' %}
|
| 120 |
+
{{- '[TOOL_RESULTS]' + message['content']|string + '[/TOOL_RESULTS]' }}
|
| 121 |
+
|
| 122 |
+
{#- Raise exception for unsupported roles. #}
|
| 123 |
+
{%- else %}
|
| 124 |
+
{{- raise_exception('Only user, assistant and tool roles are supported, got ' + message['role'] + '.') }}
|
| 125 |
+
{%- endif %}
|
| 126 |
+
{%- endfor %}
|
config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Mistral3ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"image_token_index": 10,
|
| 7 |
+
"model_type": "mistral3",
|
| 8 |
+
"multimodal_projector_bias": false,
|
| 9 |
+
"projector_hidden_act": "gelu",
|
| 10 |
+
"spatial_merge_size": 2,
|
| 11 |
+
"text_config": {
|
| 12 |
+
"attention_dropout": 0.0,
|
| 13 |
+
"head_dim": 128,
|
| 14 |
+
"hidden_act": "silu",
|
| 15 |
+
"hidden_size": 4096,
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 14336,
|
| 18 |
+
"max_position_embeddings": 262144,
|
| 19 |
+
"model_type": "ministral3",
|
| 20 |
+
"num_attention_heads": 32,
|
| 21 |
+
"num_hidden_layers": 34,
|
| 22 |
+
"num_key_value_heads": 8,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"beta_fast": 32.0,
|
| 26 |
+
"beta_slow": 1.0,
|
| 27 |
+
"factor": 16.0,
|
| 28 |
+
"llama_4_scaling_beta": 0.1,
|
| 29 |
+
"mscale": 1.0,
|
| 30 |
+
"mscale_all_dim": 1.0,
|
| 31 |
+
"original_max_position_embeddings": 16384,
|
| 32 |
+
"rope_theta": 1000000.0,
|
| 33 |
+
"rope_type": "yarn",
|
| 34 |
+
"type": "yarn"
|
| 35 |
+
},
|
| 36 |
+
"sliding_window": null,
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 131072
|
| 39 |
+
},
|
| 40 |
+
"transformers_version": "5.0.0.dev0",
|
| 41 |
+
"vision_config": {
|
| 42 |
+
"attention_dropout": 0.0,
|
| 43 |
+
"head_dim": 64,
|
| 44 |
+
"hidden_act": "silu",
|
| 45 |
+
"hidden_size": 1024,
|
| 46 |
+
"image_size": 1540,
|
| 47 |
+
"initializer_range": 0.02,
|
| 48 |
+
"intermediate_size": 4096,
|
| 49 |
+
"model_type": "pixtral",
|
| 50 |
+
"num_attention_heads": 16,
|
| 51 |
+
"num_channels": 3,
|
| 52 |
+
"num_hidden_layers": 24,
|
| 53 |
+
"patch_size": 14,
|
| 54 |
+
"rope_parameters": {
|
| 55 |
+
"rope_theta": 10000.0,
|
| 56 |
+
"rope_type": "default"
|
| 57 |
+
},
|
| 58 |
+
"rope_theta": 10000.0
|
| 59 |
+
},
|
| 60 |
+
"vision_feature_layer": -1
|
| 61 |
+
}
|
consolidated.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba5577f6f96df4b9a23addd6f49d76eb12cbc87e6c78868181d00988e9173e6e
|
| 3 |
+
size 17836115976
|
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-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6ad362f18a10e9e8dc8f1f3951ed0854aeaf427c1531a5b87a2daddf31fb2b2
|
| 3 |
+
size 4984292952
|
model-00002-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:970b5e0a46bc5f4021082b7e1087f61473d9d13cb2a483d0feb8388e2b64d251
|
| 3 |
+
size 4999804256
|
model-00003-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0b7b28ed7e8435c1867c155639fa9212a87d881752d0c4051e7c84726ebd4748
|
| 3 |
+
size 4915917680
|
model-00004-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d3f5740cd785771e26e914b1af5fb7098bceb931b89078563fb8688472c9248
|
| 3 |
+
size 2936108304
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,539 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 8918026240,
|
| 4 |
+
"total_size": 17836052480
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"language_model.lm_head.weight": "model-00004-of-00004.safetensors",
|
| 8 |
+
"language_model.model.embed_tokens.weight": "model-00001-of-00004.safetensors",
|
| 9 |
+
"language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 10 |
+
"language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 11 |
+
"language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 12 |
+
"language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 13 |
+
"language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 14 |
+
"language_model.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 15 |
+
"language_model.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 16 |
+
"language_model.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 17 |
+
"language_model.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 18 |
+
"language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 19 |
+
"language_model.model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 20 |
+
"language_model.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 21 |
+
"language_model.model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 22 |
+
"language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 23 |
+
"language_model.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 24 |
+
"language_model.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 25 |
+
"language_model.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 26 |
+
"language_model.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 27 |
+
"language_model.model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 28 |
+
"language_model.model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 29 |
+
"language_model.model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 30 |
+
"language_model.model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 31 |
+
"language_model.model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 32 |
+
"language_model.model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 33 |
+
"language_model.model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 34 |
+
"language_model.model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 35 |
+
"language_model.model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 36 |
+
"language_model.model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 37 |
+
"language_model.model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 38 |
+
"language_model.model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 39 |
+
"language_model.model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 40 |
+
"language_model.model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 41 |
+
"language_model.model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 42 |
+
"language_model.model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 43 |
+
"language_model.model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 44 |
+
"language_model.model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 45 |
+
"language_model.model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 46 |
+
"language_model.model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 47 |
+
"language_model.model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 48 |
+
"language_model.model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 49 |
+
"language_model.model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 50 |
+
"language_model.model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 51 |
+
"language_model.model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 52 |
+
"language_model.model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 53 |
+
"language_model.model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 54 |
+
"language_model.model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 55 |
+
"language_model.model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 56 |
+
"language_model.model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 57 |
+
"language_model.model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 58 |
+
"language_model.model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 59 |
+
"language_model.model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 60 |
+
"language_model.model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 61 |
+
"language_model.model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 62 |
+
"language_model.model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 63 |
+
"language_model.model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 64 |
+
"language_model.model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 65 |
+
"language_model.model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 66 |
+
"language_model.model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 67 |
+
"language_model.model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 68 |
+
"language_model.model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 69 |
+
"language_model.model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 70 |
+
"language_model.model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 71 |
+
"language_model.model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 72 |
+
"language_model.model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 73 |
+
"language_model.model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 74 |
+
"language_model.model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 75 |
+
"language_model.model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 76 |
+
"language_model.model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 77 |
+
"language_model.model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 78 |
+
"language_model.model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 79 |
+
"language_model.model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 80 |
+
"language_model.model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 81 |
+
"language_model.model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 82 |
+
"language_model.model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 83 |
+
"language_model.model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 84 |
+
"language_model.model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 85 |
+
"language_model.model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 86 |
+
"language_model.model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 87 |
+
"language_model.model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 88 |
+
"language_model.model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 89 |
+
"language_model.model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 90 |
+
"language_model.model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 91 |
+
"language_model.model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 92 |
+
"language_model.model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 93 |
+
"language_model.model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 94 |
+
"language_model.model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 95 |
+
"language_model.model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 96 |
+
"language_model.model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 97 |
+
"language_model.model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 98 |
+
"language_model.model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 99 |
+
"language_model.model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 100 |
+
"language_model.model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 101 |
+
"language_model.model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 102 |
+
"language_model.model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 103 |
+
"language_model.model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 104 |
+
"language_model.model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 105 |
+
"language_model.model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 106 |
+
"language_model.model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 107 |
+
"language_model.model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 108 |
+
"language_model.model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 109 |
+
"language_model.model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 110 |
+
"language_model.model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 111 |
+
"language_model.model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 112 |
+
"language_model.model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 113 |
+
"language_model.model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 114 |
+
"language_model.model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 115 |
+
"language_model.model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 116 |
+
"language_model.model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 117 |
+
"language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 118 |
+
"language_model.model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 119 |
+
"language_model.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 120 |
+
"language_model.model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 121 |
+
"language_model.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 122 |
+
"language_model.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 123 |
+
"language_model.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 124 |
+
"language_model.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 125 |
+
"language_model.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 126 |
+
"language_model.model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 127 |
+
"language_model.model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 128 |
+
"language_model.model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 129 |
+
"language_model.model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 130 |
+
"language_model.model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 131 |
+
"language_model.model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 132 |
+
"language_model.model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 133 |
+
"language_model.model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 134 |
+
"language_model.model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 135 |
+
"language_model.model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 136 |
+
"language_model.model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 137 |
+
"language_model.model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 138 |
+
"language_model.model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 139 |
+
"language_model.model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 140 |
+
"language_model.model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 141 |
+
"language_model.model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 142 |
+
"language_model.model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 143 |
+
"language_model.model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 144 |
+
"language_model.model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 145 |
+
"language_model.model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 146 |
+
"language_model.model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 147 |
+
"language_model.model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 148 |
+
"language_model.model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 149 |
+
"language_model.model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 150 |
+
"language_model.model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 151 |
+
"language_model.model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 152 |
+
"language_model.model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 153 |
+
"language_model.model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 154 |
+
"language_model.model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 155 |
+
"language_model.model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 156 |
+
"language_model.model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 157 |
+
"language_model.model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 158 |
+
"language_model.model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 159 |
+
"language_model.model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 160 |
+
"language_model.model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 161 |
+
"language_model.model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 162 |
+
"language_model.model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 163 |
+
"language_model.model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 164 |
+
"language_model.model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 165 |
+
"language_model.model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 166 |
+
"language_model.model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 167 |
+
"language_model.model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 168 |
+
"language_model.model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 169 |
+
"language_model.model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 170 |
+
"language_model.model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 171 |
+
"language_model.model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 172 |
+
"language_model.model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 173 |
+
"language_model.model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 174 |
+
"language_model.model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 175 |
+
"language_model.model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 176 |
+
"language_model.model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 177 |
+
"language_model.model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 178 |
+
"language_model.model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 179 |
+
"language_model.model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 180 |
+
"language_model.model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 181 |
+
"language_model.model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 182 |
+
"language_model.model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 183 |
+
"language_model.model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 184 |
+
"language_model.model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 185 |
+
"language_model.model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 186 |
+
"language_model.model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 187 |
+
"language_model.model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 188 |
+
"language_model.model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 189 |
+
"language_model.model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 190 |
+
"language_model.model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 191 |
+
"language_model.model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 192 |
+
"language_model.model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 193 |
+
"language_model.model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 194 |
+
"language_model.model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 195 |
+
"language_model.model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 196 |
+
"language_model.model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 197 |
+
"language_model.model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 198 |
+
"language_model.model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 199 |
+
"language_model.model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
|
| 200 |
+
"language_model.model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 201 |
+
"language_model.model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 202 |
+
"language_model.model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
|
| 203 |
+
"language_model.model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 204 |
+
"language_model.model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 205 |
+
"language_model.model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 206 |
+
"language_model.model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 207 |
+
"language_model.model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 208 |
+
"language_model.model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 209 |
+
"language_model.model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
|
| 210 |
+
"language_model.model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
|
| 211 |
+
"language_model.model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 212 |
+
"language_model.model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
|
| 213 |
+
"language_model.model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
|
| 214 |
+
"language_model.model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
|
| 215 |
+
"language_model.model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
|
| 216 |
+
"language_model.model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 217 |
+
"language_model.model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 218 |
+
"language_model.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 219 |
+
"language_model.model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 220 |
+
"language_model.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 221 |
+
"language_model.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 222 |
+
"language_model.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 223 |
+
"language_model.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 224 |
+
"language_model.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 225 |
+
"language_model.model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 226 |
+
"language_model.model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 227 |
+
"language_model.model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 228 |
+
"language_model.model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 229 |
+
"language_model.model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 230 |
+
"language_model.model.layers.30.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 231 |
+
"language_model.model.layers.30.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 232 |
+
"language_model.model.layers.30.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 233 |
+
"language_model.model.layers.30.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 234 |
+
"language_model.model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 235 |
+
"language_model.model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 236 |
+
"language_model.model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 237 |
+
"language_model.model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 238 |
+
"language_model.model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 239 |
+
"language_model.model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 240 |
+
"language_model.model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 241 |
+
"language_model.model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 242 |
+
"language_model.model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 243 |
+
"language_model.model.layers.32.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 244 |
+
"language_model.model.layers.32.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 245 |
+
"language_model.model.layers.32.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 246 |
+
"language_model.model.layers.32.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 247 |
+
"language_model.model.layers.32.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 248 |
+
"language_model.model.layers.32.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 249 |
+
"language_model.model.layers.32.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 250 |
+
"language_model.model.layers.32.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 251 |
+
"language_model.model.layers.32.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 252 |
+
"language_model.model.layers.33.input_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 253 |
+
"language_model.model.layers.33.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
|
| 254 |
+
"language_model.model.layers.33.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
|
| 255 |
+
"language_model.model.layers.33.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
|
| 256 |
+
"language_model.model.layers.33.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
|
| 257 |
+
"language_model.model.layers.33.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
|
| 258 |
+
"language_model.model.layers.33.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
|
| 259 |
+
"language_model.model.layers.33.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
|
| 260 |
+
"language_model.model.layers.33.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
|
| 261 |
+
"language_model.model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 262 |
+
"language_model.model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 263 |
+
"language_model.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 264 |
+
"language_model.model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 265 |
+
"language_model.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 266 |
+
"language_model.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 267 |
+
"language_model.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 268 |
+
"language_model.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 269 |
+
"language_model.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 270 |
+
"language_model.model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 271 |
+
"language_model.model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 272 |
+
"language_model.model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 273 |
+
"language_model.model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 274 |
+
"language_model.model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 275 |
+
"language_model.model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 276 |
+
"language_model.model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 277 |
+
"language_model.model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 278 |
+
"language_model.model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 279 |
+
"language_model.model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 280 |
+
"language_model.model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 281 |
+
"language_model.model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 282 |
+
"language_model.model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 283 |
+
"language_model.model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
|
| 284 |
+
"language_model.model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 285 |
+
"language_model.model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 286 |
+
"language_model.model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 287 |
+
"language_model.model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 288 |
+
"language_model.model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 289 |
+
"language_model.model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 290 |
+
"language_model.model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 291 |
+
"language_model.model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 292 |
+
"language_model.model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 293 |
+
"language_model.model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 294 |
+
"language_model.model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 295 |
+
"language_model.model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 296 |
+
"language_model.model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 297 |
+
"language_model.model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 298 |
+
"language_model.model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 299 |
+
"language_model.model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 300 |
+
"language_model.model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 301 |
+
"language_model.model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 302 |
+
"language_model.model.layers.8.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 303 |
+
"language_model.model.layers.8.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 304 |
+
"language_model.model.layers.8.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 305 |
+
"language_model.model.layers.8.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 306 |
+
"language_model.model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 307 |
+
"language_model.model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
|
| 308 |
+
"language_model.model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
|
| 309 |
+
"language_model.model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
|
| 310 |
+
"language_model.model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
|
| 311 |
+
"language_model.model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
|
| 312 |
+
"language_model.model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
|
| 313 |
+
"language_model.model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
|
| 314 |
+
"language_model.model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
|
| 315 |
+
"language_model.model.norm.weight": "model-00004-of-00004.safetensors",
|
| 316 |
+
"multi_modal_projector.linear_1.weight": "model-00001-of-00004.safetensors",
|
| 317 |
+
"multi_modal_projector.linear_2.weight": "model-00001-of-00004.safetensors",
|
| 318 |
+
"multi_modal_projector.norm.weight": "model-00001-of-00004.safetensors",
|
| 319 |
+
"multi_modal_projector.patch_merger.merging_layer.weight": "model-00001-of-00004.safetensors",
|
| 320 |
+
"vision_tower.ln_pre.weight": "model-00001-of-00004.safetensors",
|
| 321 |
+
"vision_tower.patch_conv.weight": "model-00001-of-00004.safetensors",
|
| 322 |
+
"vision_tower.transformer.layers.0.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 323 |
+
"vision_tower.transformer.layers.0.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 324 |
+
"vision_tower.transformer.layers.0.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 325 |
+
"vision_tower.transformer.layers.0.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 326 |
+
"vision_tower.transformer.layers.0.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 327 |
+
"vision_tower.transformer.layers.0.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 328 |
+
"vision_tower.transformer.layers.0.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 329 |
+
"vision_tower.transformer.layers.0.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 330 |
+
"vision_tower.transformer.layers.0.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 331 |
+
"vision_tower.transformer.layers.1.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 332 |
+
"vision_tower.transformer.layers.1.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 333 |
+
"vision_tower.transformer.layers.1.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 334 |
+
"vision_tower.transformer.layers.1.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 335 |
+
"vision_tower.transformer.layers.1.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 336 |
+
"vision_tower.transformer.layers.1.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 337 |
+
"vision_tower.transformer.layers.1.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 338 |
+
"vision_tower.transformer.layers.1.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 339 |
+
"vision_tower.transformer.layers.1.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 340 |
+
"vision_tower.transformer.layers.10.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 341 |
+
"vision_tower.transformer.layers.10.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 342 |
+
"vision_tower.transformer.layers.10.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 343 |
+
"vision_tower.transformer.layers.10.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 344 |
+
"vision_tower.transformer.layers.10.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 345 |
+
"vision_tower.transformer.layers.10.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 346 |
+
"vision_tower.transformer.layers.10.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 347 |
+
"vision_tower.transformer.layers.10.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 348 |
+
"vision_tower.transformer.layers.10.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 349 |
+
"vision_tower.transformer.layers.11.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 350 |
+
"vision_tower.transformer.layers.11.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 351 |
+
"vision_tower.transformer.layers.11.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 352 |
+
"vision_tower.transformer.layers.11.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 353 |
+
"vision_tower.transformer.layers.11.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 354 |
+
"vision_tower.transformer.layers.11.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 355 |
+
"vision_tower.transformer.layers.11.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 356 |
+
"vision_tower.transformer.layers.11.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 357 |
+
"vision_tower.transformer.layers.11.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 358 |
+
"vision_tower.transformer.layers.12.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 359 |
+
"vision_tower.transformer.layers.12.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 360 |
+
"vision_tower.transformer.layers.12.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 361 |
+
"vision_tower.transformer.layers.12.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 362 |
+
"vision_tower.transformer.layers.12.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 363 |
+
"vision_tower.transformer.layers.12.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 364 |
+
"vision_tower.transformer.layers.12.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 365 |
+
"vision_tower.transformer.layers.12.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 366 |
+
"vision_tower.transformer.layers.12.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 367 |
+
"vision_tower.transformer.layers.13.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 368 |
+
"vision_tower.transformer.layers.13.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 369 |
+
"vision_tower.transformer.layers.13.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 370 |
+
"vision_tower.transformer.layers.13.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 371 |
+
"vision_tower.transformer.layers.13.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 372 |
+
"vision_tower.transformer.layers.13.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 373 |
+
"vision_tower.transformer.layers.13.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 374 |
+
"vision_tower.transformer.layers.13.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 375 |
+
"vision_tower.transformer.layers.13.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 376 |
+
"vision_tower.transformer.layers.14.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 377 |
+
"vision_tower.transformer.layers.14.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 378 |
+
"vision_tower.transformer.layers.14.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 379 |
+
"vision_tower.transformer.layers.14.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 380 |
+
"vision_tower.transformer.layers.14.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 381 |
+
"vision_tower.transformer.layers.14.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 382 |
+
"vision_tower.transformer.layers.14.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 383 |
+
"vision_tower.transformer.layers.14.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 384 |
+
"vision_tower.transformer.layers.14.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 385 |
+
"vision_tower.transformer.layers.15.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 386 |
+
"vision_tower.transformer.layers.15.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 387 |
+
"vision_tower.transformer.layers.15.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 388 |
+
"vision_tower.transformer.layers.15.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 389 |
+
"vision_tower.transformer.layers.15.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 390 |
+
"vision_tower.transformer.layers.15.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 391 |
+
"vision_tower.transformer.layers.15.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 392 |
+
"vision_tower.transformer.layers.15.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 393 |
+
"vision_tower.transformer.layers.15.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 394 |
+
"vision_tower.transformer.layers.16.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 395 |
+
"vision_tower.transformer.layers.16.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 396 |
+
"vision_tower.transformer.layers.16.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 397 |
+
"vision_tower.transformer.layers.16.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 398 |
+
"vision_tower.transformer.layers.16.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 399 |
+
"vision_tower.transformer.layers.16.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 400 |
+
"vision_tower.transformer.layers.16.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 401 |
+
"vision_tower.transformer.layers.16.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 402 |
+
"vision_tower.transformer.layers.16.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 403 |
+
"vision_tower.transformer.layers.17.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 404 |
+
"vision_tower.transformer.layers.17.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 405 |
+
"vision_tower.transformer.layers.17.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 406 |
+
"vision_tower.transformer.layers.17.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 407 |
+
"vision_tower.transformer.layers.17.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 408 |
+
"vision_tower.transformer.layers.17.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 409 |
+
"vision_tower.transformer.layers.17.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 410 |
+
"vision_tower.transformer.layers.17.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 411 |
+
"vision_tower.transformer.layers.17.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 412 |
+
"vision_tower.transformer.layers.18.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 413 |
+
"vision_tower.transformer.layers.18.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 414 |
+
"vision_tower.transformer.layers.18.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 415 |
+
"vision_tower.transformer.layers.18.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 416 |
+
"vision_tower.transformer.layers.18.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 417 |
+
"vision_tower.transformer.layers.18.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 418 |
+
"vision_tower.transformer.layers.18.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 419 |
+
"vision_tower.transformer.layers.18.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 420 |
+
"vision_tower.transformer.layers.18.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 421 |
+
"vision_tower.transformer.layers.19.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 422 |
+
"vision_tower.transformer.layers.19.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 423 |
+
"vision_tower.transformer.layers.19.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 424 |
+
"vision_tower.transformer.layers.19.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 425 |
+
"vision_tower.transformer.layers.19.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 426 |
+
"vision_tower.transformer.layers.19.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 427 |
+
"vision_tower.transformer.layers.19.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 428 |
+
"vision_tower.transformer.layers.19.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 429 |
+
"vision_tower.transformer.layers.19.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 430 |
+
"vision_tower.transformer.layers.2.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 431 |
+
"vision_tower.transformer.layers.2.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 432 |
+
"vision_tower.transformer.layers.2.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 433 |
+
"vision_tower.transformer.layers.2.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 434 |
+
"vision_tower.transformer.layers.2.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 435 |
+
"vision_tower.transformer.layers.2.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 436 |
+
"vision_tower.transformer.layers.2.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 437 |
+
"vision_tower.transformer.layers.2.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 438 |
+
"vision_tower.transformer.layers.2.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 439 |
+
"vision_tower.transformer.layers.20.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 440 |
+
"vision_tower.transformer.layers.20.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 441 |
+
"vision_tower.transformer.layers.20.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 442 |
+
"vision_tower.transformer.layers.20.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 443 |
+
"vision_tower.transformer.layers.20.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 444 |
+
"vision_tower.transformer.layers.20.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 445 |
+
"vision_tower.transformer.layers.20.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 446 |
+
"vision_tower.transformer.layers.20.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 447 |
+
"vision_tower.transformer.layers.20.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 448 |
+
"vision_tower.transformer.layers.21.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 449 |
+
"vision_tower.transformer.layers.21.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 450 |
+
"vision_tower.transformer.layers.21.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 451 |
+
"vision_tower.transformer.layers.21.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 452 |
+
"vision_tower.transformer.layers.21.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 453 |
+
"vision_tower.transformer.layers.21.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 454 |
+
"vision_tower.transformer.layers.21.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 455 |
+
"vision_tower.transformer.layers.21.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 456 |
+
"vision_tower.transformer.layers.21.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 457 |
+
"vision_tower.transformer.layers.22.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 458 |
+
"vision_tower.transformer.layers.22.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 459 |
+
"vision_tower.transformer.layers.22.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 460 |
+
"vision_tower.transformer.layers.22.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 461 |
+
"vision_tower.transformer.layers.22.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 462 |
+
"vision_tower.transformer.layers.22.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 463 |
+
"vision_tower.transformer.layers.22.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 464 |
+
"vision_tower.transformer.layers.22.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 465 |
+
"vision_tower.transformer.layers.22.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 466 |
+
"vision_tower.transformer.layers.23.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 467 |
+
"vision_tower.transformer.layers.23.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 468 |
+
"vision_tower.transformer.layers.23.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 469 |
+
"vision_tower.transformer.layers.23.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 470 |
+
"vision_tower.transformer.layers.23.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 471 |
+
"vision_tower.transformer.layers.23.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 472 |
+
"vision_tower.transformer.layers.23.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 473 |
+
"vision_tower.transformer.layers.23.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 474 |
+
"vision_tower.transformer.layers.23.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 475 |
+
"vision_tower.transformer.layers.3.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 476 |
+
"vision_tower.transformer.layers.3.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 477 |
+
"vision_tower.transformer.layers.3.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 478 |
+
"vision_tower.transformer.layers.3.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 479 |
+
"vision_tower.transformer.layers.3.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 480 |
+
"vision_tower.transformer.layers.3.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 481 |
+
"vision_tower.transformer.layers.3.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 482 |
+
"vision_tower.transformer.layers.3.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 483 |
+
"vision_tower.transformer.layers.3.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 484 |
+
"vision_tower.transformer.layers.4.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 485 |
+
"vision_tower.transformer.layers.4.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 486 |
+
"vision_tower.transformer.layers.4.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 487 |
+
"vision_tower.transformer.layers.4.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 488 |
+
"vision_tower.transformer.layers.4.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 489 |
+
"vision_tower.transformer.layers.4.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 490 |
+
"vision_tower.transformer.layers.4.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 491 |
+
"vision_tower.transformer.layers.4.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 492 |
+
"vision_tower.transformer.layers.4.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 493 |
+
"vision_tower.transformer.layers.5.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 494 |
+
"vision_tower.transformer.layers.5.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 495 |
+
"vision_tower.transformer.layers.5.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 496 |
+
"vision_tower.transformer.layers.5.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 497 |
+
"vision_tower.transformer.layers.5.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 498 |
+
"vision_tower.transformer.layers.5.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 499 |
+
"vision_tower.transformer.layers.5.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 500 |
+
"vision_tower.transformer.layers.5.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 501 |
+
"vision_tower.transformer.layers.5.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 502 |
+
"vision_tower.transformer.layers.6.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 503 |
+
"vision_tower.transformer.layers.6.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 504 |
+
"vision_tower.transformer.layers.6.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 505 |
+
"vision_tower.transformer.layers.6.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 506 |
+
"vision_tower.transformer.layers.6.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 507 |
+
"vision_tower.transformer.layers.6.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 508 |
+
"vision_tower.transformer.layers.6.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 509 |
+
"vision_tower.transformer.layers.6.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 510 |
+
"vision_tower.transformer.layers.6.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 511 |
+
"vision_tower.transformer.layers.7.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 512 |
+
"vision_tower.transformer.layers.7.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 513 |
+
"vision_tower.transformer.layers.7.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 514 |
+
"vision_tower.transformer.layers.7.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 515 |
+
"vision_tower.transformer.layers.7.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 516 |
+
"vision_tower.transformer.layers.7.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 517 |
+
"vision_tower.transformer.layers.7.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 518 |
+
"vision_tower.transformer.layers.7.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 519 |
+
"vision_tower.transformer.layers.7.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 520 |
+
"vision_tower.transformer.layers.8.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 521 |
+
"vision_tower.transformer.layers.8.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 522 |
+
"vision_tower.transformer.layers.8.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 523 |
+
"vision_tower.transformer.layers.8.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 524 |
+
"vision_tower.transformer.layers.8.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 525 |
+
"vision_tower.transformer.layers.8.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 526 |
+
"vision_tower.transformer.layers.8.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 527 |
+
"vision_tower.transformer.layers.8.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 528 |
+
"vision_tower.transformer.layers.8.ffn_norm.weight": "model-00001-of-00004.safetensors",
|
| 529 |
+
"vision_tower.transformer.layers.9.attention.k_proj.weight": "model-00001-of-00004.safetensors",
|
| 530 |
+
"vision_tower.transformer.layers.9.attention.o_proj.weight": "model-00001-of-00004.safetensors",
|
| 531 |
+
"vision_tower.transformer.layers.9.attention.q_proj.weight": "model-00001-of-00004.safetensors",
|
| 532 |
+
"vision_tower.transformer.layers.9.attention.v_proj.weight": "model-00001-of-00004.safetensors",
|
| 533 |
+
"vision_tower.transformer.layers.9.attention_norm.weight": "model-00001-of-00004.safetensors",
|
| 534 |
+
"vision_tower.transformer.layers.9.feed_forward.down_proj.weight": "model-00001-of-00004.safetensors",
|
| 535 |
+
"vision_tower.transformer.layers.9.feed_forward.gate_proj.weight": "model-00001-of-00004.safetensors",
|
| 536 |
+
"vision_tower.transformer.layers.9.feed_forward.up_proj.weight": "model-00001-of-00004.safetensors",
|
| 537 |
+
"vision_tower.transformer.layers.9.ffn_norm.weight": "model-00001-of-00004.safetensors"
|
| 538 |
+
}
|
| 539 |
+
}
|
params.json
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"dim": 4096,
|
| 3 |
+
"n_layers": 34,
|
| 4 |
+
"head_dim": 128,
|
| 5 |
+
"hidden_dim": 14336,
|
| 6 |
+
"n_heads": 32,
|
| 7 |
+
"n_kv_heads": 8,
|
| 8 |
+
"rope_theta": 1000000.0,
|
| 9 |
+
"norm_eps": 1e-05,
|
| 10 |
+
"vocab_size": 131072,
|
| 11 |
+
"tied_embeddings": false,
|
| 12 |
+
"max_position_embeddings": 262144,
|
| 13 |
+
"llama_4_scaling": {
|
| 14 |
+
"original_max_position_embeddings": 16384,
|
| 15 |
+
"beta": 0.1
|
| 16 |
+
},
|
| 17 |
+
"q_lora_rank": null,
|
| 18 |
+
"qk_rope_head_dim": null,
|
| 19 |
+
"qk_nope_head_dim": null,
|
| 20 |
+
"kv_lora_rank": null,
|
| 21 |
+
"v_head_dim": null,
|
| 22 |
+
"yarn": {
|
| 23 |
+
"original_max_position_embeddings": 16384,
|
| 24 |
+
"factor": 16,
|
| 25 |
+
"apply_scale": false,
|
| 26 |
+
"beta": 32,
|
| 27 |
+
"alpha": 1
|
| 28 |
+
},
|
| 29 |
+
"vision_encoder": {
|
| 30 |
+
"image_token_id": 10,
|
| 31 |
+
"intermediate_size": 4096,
|
| 32 |
+
"num_hidden_layers": 24,
|
| 33 |
+
"num_attention_heads": 16,
|
| 34 |
+
"mm_projector_id": "patch_merge",
|
| 35 |
+
"spatial_merge_size": 2,
|
| 36 |
+
"hidden_size": 1024,
|
| 37 |
+
"num_channels": 3,
|
| 38 |
+
"image_size": 1540,
|
| 39 |
+
"max_image_size": 1540,
|
| 40 |
+
"patch_size": 14,
|
| 41 |
+
"rope_theta": 10000.0,
|
| 42 |
+
"add_pre_mm_projector_layer_norm": true,
|
| 43 |
+
"adapter_bias": false,
|
| 44 |
+
"image_break_token_id": 12,
|
| 45 |
+
"image_end_token_id": 13
|
| 46 |
+
}
|
| 47 |
+
}
|
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 |
+
}
|
special_tokens_map.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tekken.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e29d19ea32eb7e26e6c0572d57cb7f9eca0f4420e0e0fe6ae1cf3be94da1c0d6
|
| 3 |
+
size 16753777
|
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
|
The diff for this file is too large to render.
See raw diff
|
|
|