EllieS
/

zephyr-7b-dpo-lora-pubmedqa-mix2

alignment-handbook

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

zephyr-7b-dpo-lora-pubmedqa-mix2 / train_results.json

EllieS's picture

Model save

8f17533 verified 8 months ago

198 Bytes

	{
	"epoch": 1.0,
	"train_loss": 0.05805659896943479,
	"train_runtime": 136192.1568,
	"train_samples": 334157,
	"train_samples_per_second": 2.454,
	"train_steps_per_second": 0.613
	}