YYYYYYibo
/

zephyr-7b-dpo-qlora-min-pi-part-0

alignment-handbook

Generated from Trainer

Model card Files Files and versions Community

zephyr-7b-dpo-qlora-min-pi-part-0 / train_results.json

YYYYYYibo's picture

Model save

4738e8b verified 6 months ago

No virus

193 Bytes

	{
	"epoch": 0.77,
	"train_loss": 0.692988375822703,
	"train_runtime": 257.3561,
	"train_samples": 10000,
	"train_samples_per_second": 3.886,
	"train_steps_per_second": 0.012
	}