DUAL-GPO
/

zephyr-7b-gpo-update3-i0

alignment-handbook

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

zephyr-7b-gpo-update3-i0 / train_results.json

lole25's picture

Model save

0174c0e verified 2 months ago

raw history blame contribute delete

No virus

198 Bytes

	{
	"epoch": 1.0,
	"train_loss": 0.028396672180453324,
	"train_runtime": 172620.3313,
	"train_samples": 61135,
	"train_samples_per_second": 0.354,
	"train_steps_per_second": 0.089
	}