YYYYYYibo
/

nash_dpo_doff_real_no_golden_iter_1

alignment-handbook

Generated from Trainer

Model card Files Files and versions Community

nash_dpo_doff_real_no_golden_iter_1 / all_results.json

YYYYYYibo's picture

Model save

570dd6b verified 8 months ago

192 Bytes

	{
	"epoch": 1.0,
	"train_loss": 0.682184520889731,
	"train_runtime": 6549.791,
	"train_samples": 19600,
	"train_samples_per_second": 2.992,
	"train_steps_per_second": 0.023
	}