lole25
/

phi-2-sft-lora-ultrachat

alignment-handbook

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

phi-2-sft-lora-ultrachat / eval_results.json

lole25's picture

Model save

bd11305 verified 8 months ago

190 Bytes

	{
	"epoch": 3.0,
	"eval_loss": 1.2043143510818481,
	"eval_runtime": 1766.7333,
	"eval_samples": 23110,
	"eval_samples_per_second": 16.022,
	"eval_steps_per_second": 1.002
	}