EllieS
/

zephyr-7b-dpo-lora-pubmedqa-selfgen-old

alignment-handbook

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

zephyr-7b-dpo-lora-pubmedqa-selfgen-old / train_results.json

EllieS's picture

Model save

bf8ff8e verified 9 months ago

197 Bytes

	{
	"epoch": 1.0,
	"train_loss": 0.03956493256323719,
	"train_runtime": 68990.3604,
	"train_samples": 134157,
	"train_samples_per_second": 1.945,
	"train_steps_per_second": 0.243
	}