Instructions to use stojchet/lr_kto2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use stojchet/lr_kto2 with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
Download params.json from stojchet/lr_kto2: direct link, hf CLI and curl.
- Browser
- Download file 373 Bytes
-
https://huggingface.co/stojchet/lr_kto2/resolve/main/params.json
- Command line
-
hf download hf://stojchet/lr_kto2/params.json
-
curl -L -o params.json https://huggingface.co/stojchet/lr_kto2/resolve/main/params.json
373 Bytes
| {"max_seq_length": 1500, "language": "python", "dataset_size": 10000, "epochs": 1, "per_device_train_batch_size": 8, "gradient_accumulation_steps": 16, "learning_rate": 1.41e-06, "lora_r": 64, "lora_alpha": 16, "lora_dropout": 0.05, "warmup_ratio": 0.1, "dataset_name": "stojchet/kto-deepseek-coder-1.3b-base-empty-fn", "base_model": "deepseek-ai/deepseek-coder-1.3b-base"} |