run3_low_lr
QLoRA adapter fine-tuned from Qwen/Qwen2.5-3B-Instruct on OpenAssistant/oasst1.
Final metrics
- train_loss: 1.325200669419268
- eval_loss: 1.4974315166473389
- loss_gap: 0.17223084722807092
- eval_perplexity: 4.470192697011084
- train_runtime: 3985.3134
- samples_per_second: 6.818
- global_steps: 1701
- Downloads last month
- 7