WL2928
/

zephyr-7b-dpo-qlora

alignment-handbook

Generated from Trainer

4-bit precision

Model card Files Files and versions Metrics Training metrics Community

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

WL2928's picture

Model save

4398fcf verified 3 months ago

raw history blame contribute delete

No virus

196 Bytes

	{
	"epoch": 1.0,
	"train_loss": 0.5201432038994555,
	"train_runtime": 240641.1375,
	"train_samples": 61135,
	"train_samples_per_second": 0.254,
	"train_steps_per_second": 0.032
	}