just1nseo
/

zephyr-dpop-qlora-gpt4-5e-7-epoch3

alignment-handbook

Generated from Trainer

4-bit precision

Model card Files Files and versions Metrics Training metrics Community

zephyr-dpop-qlora-gpt4-5e-7-epoch3 / train_results.json

just1nseo's picture

Model save

bf9d83a verified 5 months ago

194 Bytes

	{
	"epoch": 3.0,
	"train_loss": 0.5596254680078354,
	"train_runtime": 11165.5936,
	"train_samples": 5678,
	"train_samples_per_second": 1.526,
	"train_steps_per_second": 0.095
	}