sengi
/

zephyr-7b-pl-qlora

alignment-handbook

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

zephyr-7b-pl-qlora / train_results.json

sengi's picture

Model save

66b2cbf verified 7 months ago

198 Bytes

	{
	"epoch": 0.28,
	"train_loss": 0.09296217595246466,
	"train_runtime": 26220.1068,
	"train_samples": 207865,
	"train_samples_per_second": 1.465,
	"train_steps_per_second": 0.092
	}