abgoswam
/

zephyr-7b-gemma-dpo

Text Generation

alignment-handbook

Generated from Trainer

text-generation-inference

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

zephyr-7b-gemma-dpo / train_results.json

abgoswam's picture

Model save

dbcdf3d verified 11 months ago

196 Bytes

	{
	"epoch": 1.97,
	"train_loss": 0.38778070876231563,
	"train_runtime": 1146.4695,
	"train_samples": 6750,
	"train_samples_per_second": 11.775,
	"train_steps_per_second": 0.091
	}