lewtun
/

gemma-7b-dpo-full-mix1-beta-0.1-epoch-3

Text Generation

alignment-handbook

generated_from_trainer

Inference Endpoints

text-generation-inference

Model card Files Files and versions Community

gemma-7b-dpo-full-mix1-beta-0.1-epoch-3 / eval_results.json

lewtun's picture

lewtun HF staff

End of training

edb0f47 verified 3 months ago

raw history blame contribute delete

No virus

554 Bytes

	{
	"epoch": 2.96,
	"eval_logits/chosen": 100.24970245361328,
	"eval_logits/rejected": 94.2728271484375,
	"eval_logps/chosen": -485.094482421875,
	"eval_logps/rejected": -504.0652160644531,
	"eval_loss": 0.5433239936828613,
	"eval_rewards/accuracies": 0.71875,
	"eval_rewards/chosen": -3.1689956188201904,
	"eval_rewards/margins": 2.082571268081665,
	"eval_rewards/rejected": -5.2515668869018555,
	"eval_runtime": 49.158,
	"eval_samples": 750,
	"eval_samples_per_second": 15.257,
	"eval_steps_per_second": 0.488
	}