Instructions to use timlim123/zephyr-7b-sft-qlora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use timlim123/zephyr-7b-sft-qlora with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-v0.1") model = PeftModel.from_pretrained(base_model, "timlim123/zephyr-7b-sft-qlora") - Notebooks
- Google Colab
- Kaggle
Download all_results.json from timlim123/zephyr-7b-sft-qlora: direct link, hf CLI and curl.
- Browser
- Download file 197 Bytes
-
https://huggingface.co/timlim123/zephyr-7b-sft-qlora/resolve/main/all_results.json
- Command line
-
hf download hf://timlim123/zephyr-7b-sft-qlora/all_results.json
-
curl -L -o all_results.json https://huggingface.co/timlim123/zephyr-7b-sft-qlora/resolve/main/all_results.json
197 Bytes
| { | |
| "epoch": 1.0, | |
| "train_loss": 0.9532137394142939, | |
| "train_runtime": 251062.1903, | |
| "train_samples": 207864, | |
| "train_samples_per_second": 0.555, | |
| "train_steps_per_second": 0.069 | |
| } |