Instructions to use Ex0bit/jit-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use Ex0bit/jit-lora with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir jit-lora Ex0bit/jit-lora
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- LM Studio
| { | |
| "framework": "mlx", | |
| "model_type": "jit-lora", | |
| "lora_config": { | |
| "rank": 32, | |
| "alpha": 32.0, | |
| "targets": [ | |
| "q_proj", | |
| "v_proj", | |
| "out_proj", | |
| "down_proj" | |
| ], | |
| "dropout": 0.0 | |
| }, | |
| "training": { | |
| "backend": "mlx", | |
| "learning_rate": 0.0005, | |
| "max_seq_len": 512, | |
| "batch_size": 0, | |
| "epochs": 15 | |
| } | |
| } |