awsuineg
/

zephyr-orpo-7b-hehe

Text Generation

Generated from Trainer

alignment-handbook

text-generation-inference

Inference Endpoints

4-bit precision

Model card Files Files and versions Metrics Training metrics Community

zephyr-orpo-7b-hehe / train_results.json

awsuineg's picture

Model save

3c9a152 verified 3 months ago

215 Bytes

	{
	"epoch": 3.0,
	"total_flos": 0.0,
	"train_loss": 1.4294498003771459,
	"train_runtime": 7808.9602,
	"train_samples": 7210,
	"train_samples_per_second": 2.77,
	"train_steps_per_second": 0.173
	}