|
--- |
|
license: mit |
|
library_name: transformers |
|
base_model: deepseek-ai/DeepSeek-R1-Distill-Llama-8B |
|
tags: |
|
- mlx |
|
--- |
|
|
|
# decisionslab/Dlab-852-4bit |
|
|
|
The Model [decisionslab/Dlab-852-4bit](https://huggingface.co/decisionslab/Dlab-852-4bit) was |
|
converted to MLX format from [deepseek-ai/DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) |
|
using mlx-lm version **0.21.1**. |
|
|
|
## Use with mlx |
|
|
|
```bash |
|
pip install mlx-lm |
|
``` |
|
|
|
```python |
|
from mlx_lm import load, generate |
|
|
|
model, tokenizer = load("decisionslab/Dlab-852-8B-4bit") |
|
|
|
prompt = "hello" |
|
|
|
if tokenizer.chat_template is not None: |
|
messages = [{"role": "user", "content": prompt}] |
|
prompt = tokenizer.apply_chat_template( |
|
messages, add_generation_prompt=True |
|
) |
|
|
|
response = generate(model, tokenizer, prompt=prompt, verbose=True) |
|
``` |
|
|
|
|
|
# decisionslab/Dlab-852-4bit |
|
|
|
The Model [decisionslab/Dlab-852-4bit](https://huggingface.co/decisionslab/Dlab-852-4bit) was |
|
converted to MLX format from [deepseek-ai/DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) |
|
using mlx-lm version **0.21.1**. |
|
|
|
|
|
## Model Overview |
|
|
|
**Model Name:**[decisionslab/Dlab-852-4bit](https://huggingface.co/decisionslab/Dlab-852-4bit) |
|
**Base Model:** [deepseek-ai/DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) |
|
**Intended Use:** Cultural aligned deep reasoning for Hong Kong |
|
**Language(s):** Primarily English |
|
|
|
## Model Description |
|
|
|
[decisionslab/Dlab-852-4bit](https://huggingface.co/decisionslab/Dlab-852-4bit) is a fine-tuned version of [deepseek-ai/DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B), optimized to align with the cultural and social perspectives relevant to Hong Kong. The model is trained using a dataset that includes World Values Survey data and additional Hong Kong-specific datasets curated by Decisions Lab. The goal of this fine-tuning process is to enhance the model's capability with cultural alignment to Hong Kong for deep thinking, and contextual reasoning. |
|
|
|
## Intended Use Cases |
|
|
|
- Policy simulation and decision support in Hong Kong-related contexts. |
|
- Deep reasoning tasks involving multi-perspective analysis. |
|
- Language and social interaction modeling tailored for Hong Kong users. |
|
|
|
## Evaluation |
|
|
|
The model is currently under evaluation, and CD Eval results will be published in a future update. |
|
|
|
## License |
|
|
|
All content in this repository is proprietary and confidential. The software and any associated documentation files are the exclusive property of Decisions Lab. Unauthorized copying, distribution, modification, or use of this software, via any medium, is strictly prohibited. Access to and use of this software requires explicit permission from Decisions Lab. |
|
|
|
© 2025 Decisions Lab. All rights reserved. |
|
|
|
## Contact |
|
|
|
For inquiries, collaborations, or feedback, please contact **Decisions Lab** via [[email protected]](mailto:[email protected]). |
