|
--- |
|
language: |
|
- code |
|
license: llama2 |
|
tags: |
|
- llama-2 |
|
- mlx |
|
- mlx |
|
pipeline_tag: text-generation |
|
--- |
|
|
|
# mlx-community/CodelLama7B-inst-dpo-7k-mlx |
|
This model is a finetuned MLX version from [`mlx-community/CodeLlama-7b-Instruct-hf-4bit-MLX`](). |
|
Refer to the [original model card](https://huggingface.co/mlx-community/CodeLlama-7b-Instruct-hf-4bit-MLX) for more details on the model. |
|
|
|
Using argilla 7k DPO dataset |
|
## Use with mlx |
|
|
|
```bash |
|
pip install mlx-lm |
|
``` |
|
|
|
```python |
|
from mlx_lm import load, generate |
|
|
|
model, tokenizer = load("mlx-community/CodelLama7B-inst-dpo-7k-mlx") |
|
response = generate(model, tokenizer, prompt="hello", verbose=True) |
|
``` |
|
|
