all_results.json

{
 "epoch": 1.0,
 "eval_logits/chosen": -1.3358324766159058,
 "eval_logits/rejected": -1.3033332824707031,
 "eval_logps/chosen": -2612.982421875,
 "eval_logps/rejected": -2212.504638671875,
 "eval_loss": 0.6284635066986084,
 "eval_rewards/accuracies": 0.6620000004768372,
 "eval_rewards/chosen": 0.5979290008544922,
 "eval_rewards/margins": 0.41780921816825867,
 "eval_rewards/rejected": 0.1801198571920395,
 "eval_runtime": 276.5167,
 "eval_samples": 2000,
 "eval_samples_per_second": 7.233,
 "eval_steps_per_second": 0.452,
 "train_loss": 0.6510803117330272,
 "train_runtime": 30377.0855,
 "train_samples": 61135,
 "train_samples_per_second": 2.013,
 "train_steps_per_second": 0.126
}