Edit model card
Training procedure
Framework versions
- eval_loss = 0.193117
- PEFT 0.4.0
- learning_rate = 1e-5
- lora_r = 16
- lora_alpha = 64
- lora_dropout = 0.05
- gradient_accumulation_steps = 8
- per_device_train_batch_size = 8
- optim = "paged_adamw_8bit"
Unable to determine this model’s pipeline type. Check the
docs
.