David-Xu
/

cira-7b-dpo-lora-merge

alignment-handbook

Generated from Trainer

4-bit precision

Model card Files Files and versions Metrics Training metrics Community

cira-7b-dpo-lora-merge / train_results.json

David-Xu's picture

Training in progress, step 900

89bd182 verified 8 months ago

193 Bytes

	{
	"epoch": 1.0,
	"train_loss": 0.06080360662445443,
	"train_runtime": 395.6281,
	"train_samples": 3588,
	"train_samples_per_second": 9.069,
	"train_steps_per_second": 2.267
	}