Instructions to use aamijar/correct_init_BA_SVD with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use aamijar/correct_init_BA_SVD with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("aamijar/correct_init_BA_SVD", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download adapter_config.json from aamijar/correct_init_BA_SVD: direct link, hf CLI and curl.
- Browser
- Download file 483 Bytes
-
https://huggingface.co/aamijar/correct_init_BA_SVD/resolve/main/adapter_config.json
- Command line
-
hf download hf://aamijar/correct_init_BA_SVD/adapter_config.json
-
curl -L -o adapter_config.json https://huggingface.co/aamijar/correct_init_BA_SVD/resolve/main/adapter_config.json
483 Bytes
| { | |
| "auto_mapping": null, | |
| "base_model_name_or_path": "aamijar/shortened-llm-Llama-2-7b-hf-taylor-n10-rm-6-blocks", | |
| "inference_mode": true, | |
| "init_weights": true, | |
| "layers_to_transform": null, | |
| "lora_alpha": 16, | |
| "lora_dropout": 0.1, | |
| "peft_type": "PORTLORA", | |
| "r": 4096, | |
| "revision": null, | |
| "target_modules": [ | |
| "q_proj", | |
| "v_proj", | |
| "up_proj", | |
| "down_proj", | |
| "gate_proj", | |
| "k_proj", | |
| "o_proj" | |
| ], | |
| "task_type": "CAUSAL_LM", | |
| "use_rslora": false | |
| } |