just1nseo
/

zephyr-7b-dpo-full-accumulation4

Text Generation

alignment-handbook

Generated from Trainer

Inference Endpoints

text-generation-inference

Model card Files Files and versions Community

zephyr-7b-dpo-full-accumulation4 / train_results.json

just1nseo's picture

Model save

c39a015 verified 2 months ago

raw history blame contribute delete

No virus

233 Bytes

	{
	"epoch": 0.9984301412872841,
	"total_flos": 0.0,
	"train_loss": 0.5498417743346976,
	"train_runtime": 10726.7175,
	"train_samples": 61134,
	"train_samples_per_second": 5.699,
	"train_steps_per_second": 0.044
	}