DUAL-GPO
/

phi-2-gpo-test-longest-iter-4

alignment-handbook

Generated from Trainer

Model card Files Files and versions Community

phi-2-gpo-test-longest-iter-4 / train_results.json

BraylonDash's picture

Model save

2974093 verified 3 months ago

raw history blame contribute delete

No virus

197 Bytes

	{
	"epoch": 1.98,
	"train_loss": 0.010321829197627882,
	"train_runtime": 1406.1718,
	"train_samples": 61135,
	"train_samples_per_second": 1.422,
	"train_steps_per_second": 0.088
	}