Minbyul
/

llama2-7b-dpo-full-sft-wo-medication_qa

Text Generation

alignment-handbook

Generated from Trainer

Inference Endpoints

text-generation-inference

Model card Files Files and versions Community

llama2-7b-dpo-full-sft-wo-medication_qa / eval_results.json

Minbyul's picture

End of training

258753f verified about 2 months ago

raw history blame contribute delete

No virus

571 Bytes

	{
	"epoch": 1.0,
	"eval_logits/chosen": -0.42766323685646057,
	"eval_logits/rejected": -0.4719906747341156,
	"eval_logps/chosen": -477.8221130371094,
	"eval_logps/rejected": -650.3414306640625,
	"eval_loss": 0.4395619034767151,
	"eval_rewards/accuracies": 0.949999988079071,
	"eval_rewards/chosen": -0.17786642909049988,
	"eval_rewards/margins": 1.0689051151275635,
	"eval_rewards/rejected": -1.2467713356018066,
	"eval_runtime": 37.8151,
	"eval_samples": 320,
	"eval_samples_per_second": 8.462,
	"eval_steps_per_second": 0.264
	}