Minbyul
/

meditron-7b-dpo-full-sft-wo-healthsearch_qa

Text Generation

alignment-handbook

Generated from Trainer

Inference Endpoints

text-generation-inference

Model card Files Files and versions Community

meditron-7b-dpo-full-sft-wo-healthsearch_qa / eval_results.json

Minbyul's picture

End of training

6afa495 verified about 2 months ago

raw history blame contribute delete

No virus

575 Bytes

	{
	"epoch": 1.0,
	"eval_logits/chosen": -0.8803853392601013,
	"eval_logits/rejected": -0.776340126991272,
	"eval_logps/chosen": -564.5243530273438,
	"eval_logps/rejected": -893.765869140625,
	"eval_loss": 0.6910878419876099,
	"eval_rewards/accuracies": 0.656593382358551,
	"eval_rewards/chosen": 0.0010129850124940276,
	"eval_rewards/margins": 0.0048895603977143764,
	"eval_rewards/rejected": -0.003876575967296958,
	"eval_runtime": 293.2513,
	"eval_samples": 2886,
	"eval_samples_per_second": 9.841,
	"eval_steps_per_second": 0.31
	}