{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9874476987447699,
  "eval_steps": 500,
  "global_step": 59,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.016736401673640166,
      "grad_norm": 7.77175439589368,
      "learning_rate": 8.333333333333333e-08,
      "logits/chosen": -2.807276487350464,
      "logits/rejected": -2.7759768962860107,
      "logps/chosen": -315.42626953125,
      "logps/rejected": -227.5915985107422,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16736401673640167,
      "grad_norm": 7.460865802174484,
      "learning_rate": 4.930057285201027e-07,
      "logits/chosen": -2.7554690837860107,
      "logits/rejected": -2.7460832595825195,
      "logps/chosen": -271.9754638671875,
      "logps/rejected": -260.6775207519531,
      "loss": 0.6915,
      "rewards/accuracies": 0.5104166865348816,
      "rewards/chosen": 0.00494508258998394,
      "rewards/margins": 0.0031111964490264654,
      "rewards/rejected": 0.0018338869558647275,
      "step": 10
    },
    {
      "epoch": 0.33472803347280333,
      "grad_norm": 6.716729624177502,
      "learning_rate": 4.187457503795526e-07,
      "logits/chosen": -2.786860704421997,
      "logits/rejected": -2.7680721282958984,
      "logps/chosen": -261.44482421875,
      "logps/rejected": -250.84896850585938,
      "loss": 0.6746,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": 0.03236861526966095,
      "rewards/margins": 0.04043348878622055,
      "rewards/rejected": -0.008064868859946728,
      "step": 20
    },
    {
      "epoch": 0.502092050209205,
      "grad_norm": 7.411189448055284,
      "learning_rate": 2.8691164100062034e-07,
      "logits/chosen": -2.7879672050476074,
      "logits/rejected": -2.768947124481201,
      "logps/chosen": -293.8474426269531,
      "logps/rejected": -254.21304321289062,
      "loss": 0.6486,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.02391352318227291,
      "rewards/margins": 0.12663637101650238,
      "rewards/rejected": -0.15054988861083984,
      "step": 30
    },
    {
      "epoch": 0.6694560669456067,
      "grad_norm": 8.251416788282627,
      "learning_rate": 1.4248369943086995e-07,
      "logits/chosen": -2.7608864307403564,
      "logits/rejected": -2.7383551597595215,
      "logps/chosen": -264.16790771484375,
      "logps/rejected": -255.6016387939453,
      "loss": 0.6353,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.11248473078012466,
      "rewards/margins": 0.160425066947937,
      "rewards/rejected": -0.2729097902774811,
      "step": 40
    },
    {
      "epoch": 0.8368200836820083,
      "grad_norm": 8.571363712845116,
      "learning_rate": 3.473909705816111e-08,
      "logits/chosen": -2.770698070526123,
      "logits/rejected": -2.7507286071777344,
      "logps/chosen": -279.0198059082031,
      "logps/rejected": -287.19866943359375,
      "loss": 0.6201,
      "rewards/accuracies": 0.621874988079071,
      "rewards/chosen": -0.22011137008666992,
      "rewards/margins": 0.13675080239772797,
      "rewards/rejected": -0.3568621575832367,
      "step": 50
    },
    {
      "epoch": 0.9874476987447699,
      "step": 59,
      "total_flos": 0.0,
      "train_loss": 0.6490308228185622,
      "train_runtime": 1635.4122,
      "train_samples_per_second": 9.345,
      "train_steps_per_second": 0.036
    }
  ],
  "logging_steps": 10,
  "max_steps": 59,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}