PEFT
Safetensors
mistral
alignment-handbook
trl
sft
unsloth
Generated from Trainer
4-bit precision
bitsandbytes
Instructions to use Peter/uploadtestsmallstep2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Peter/uploadtestsmallstep2 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("unsloth/mistral-7b") model = PeftModel.from_pretrained(base_model, "Peter/uploadtestsmallstep2") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Unsloth Desktop
Download all_results.json from Peter/uploadtestsmallstep2: direct link, hf CLI and curl.
- Browser
- Download file 359 Bytes
-
https://huggingface.co/Peter/uploadtestsmallstep2/resolve/fab2a5e365b0e6b0c5eb45f226937f0ecdd780c2/all_results.json
- Command line
-
hf download hf://Peter/uploadtestsmallstep2@fab2a5e365b0e6b0c5eb45f226937f0ecdd780c2/all_results.json
-
curl -L -o all_results.json https://huggingface.co/Peter/uploadtestsmallstep2/resolve/fab2a5e365b0e6b0c5eb45f226937f0ecdd780c2/all_results.json
359 Bytes
| { | |
| "epoch": 0.03, | |
| "eval_loss": 0.3438417911529541, | |
| "eval_runtime": 87.8016, | |
| "eval_samples": 249, | |
| "eval_samples_per_second": 2.836, | |
| "eval_steps_per_second": 0.364, | |
| "train_loss": 0.43747174739837646, | |
| "train_runtime": 462.0146, | |
| "train_samples": 2854, | |
| "train_samples_per_second": 0.173, | |
| "train_steps_per_second": 0.011 | |
| } |