Instructions to use Sameh515/Shalabi-1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use Sameh515/Shalabi-1 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download Sameh515/Shalabi-1 --local-dir Shalabi-1
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download task_config.json from Sameh515/Shalabi-1: direct link, hf CLI and curl.
- Browser
- Download file 736 Bytes
-
https://huggingface.co/Sameh515/Shalabi-1/resolve/main/task_config.json
- Command line
-
hf download hf://Sameh515/Shalabi-1/task_config.json
-
curl -L -o task_config.json https://huggingface.co/Sameh515/Shalabi-1/resolve/main/task_config.json
736 Bytes
| { | |
| "name": "Next Step v0.1", | |
| "base_model": "mlx-community/Qwen3.5-9B-4bit", | |
| "base_revision": "8b2b98c00a6b4d291155e4890773ca8f769aee53", | |
| "max_input_tokens": 256, | |
| "seed": 42, | |
| "num_layers": 1, | |
| "lora_parameters": { | |
| "rank": 8, | |
| "scale": 20.0, | |
| "dropout": 0.0, | |
| "keys": [ | |
| "self_attn.q_proj", | |
| "self_attn.v_proj" | |
| ] | |
| }, | |
| "learning_rate": 5e-05, | |
| "updates": 270, | |
| "validation_every": 90, | |
| "expected_trainable_parameters": 139264, | |
| "forward_mode": "eval", | |
| "scores_atol": 0.0001, | |
| "memory_limit_gb": 32, | |
| "cache_limit_gb": 2, | |
| "bootstrap_samples": 10000, | |
| "author": "Sameh Shalabi", | |
| "author_url": "https://www.linkedin.com/in/sam-shalabi/", | |
| "repo_slug": "qwen3.5-9b-next-step-lora-v0.1" | |
| } | |