{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.888888888888889,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 5.219141483306885,
      "learning_rate": 0.00018,
      "logits/chosen": -0.2618408203125,
      "logits/rejected": -0.405269056558609,
      "logps/chosen": -146.83187866210938,
      "logps/rejected": -130.77682495117188,
      "loss": 0.6931,
      "rewards/accuracies": 0.42500001192092896,
      "rewards/chosen": 0.024159394204616547,
      "rewards/margins": 0.020402083173394203,
      "rewards/rejected": 0.0037573135923594236,
      "step": 10
    },
    {
      "epoch": 1.1777777777777778,
      "grad_norm": 6.610161781311035,
      "learning_rate": 0.00019333333333333333,
      "logits/chosen": 0.21812255680561066,
      "logits/rejected": 0.1329454481601715,
      "logps/chosen": -144.95765686035156,
      "logps/rejected": -133.71145629882812,
      "loss": 0.682,
      "rewards/accuracies": 0.594936728477478,
      "rewards/chosen": -0.0013131718151271343,
      "rewards/margins": 0.18159937858581543,
      "rewards/rejected": -0.18291252851486206,
      "step": 20
    },
    {
      "epoch": 1.7703703703703704,
      "grad_norm": 2.0350332260131836,
      "learning_rate": 0.00018500000000000002,
      "logits/chosen": 0.40491190552711487,
      "logits/rejected": 0.2921258509159088,
      "logps/chosen": -140.00352478027344,
      "logps/rejected": -134.99171447753906,
      "loss": 0.3351,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 0.19238564372062683,
      "rewards/margins": 1.1973682641983032,
      "rewards/rejected": -1.004982590675354,
      "step": 30
    },
    {
      "epoch": 2.3555555555555556,
      "grad_norm": 2.579374313354492,
      "learning_rate": 0.00017666666666666666,
      "logits/chosen": 0.11085856705904007,
      "logits/rejected": 0.032480403780937195,
      "logps/chosen": -143.1613006591797,
      "logps/rejected": -159.1355438232422,
      "loss": 0.1806,
      "rewards/accuracies": 0.9620253443717957,
      "rewards/chosen": 0.3689553439617157,
      "rewards/margins": 2.494295358657837,
      "rewards/rejected": -2.125339984893799,
      "step": 40
    },
    {
      "epoch": 2.948148148148148,
      "grad_norm": 0.8431211709976196,
      "learning_rate": 0.00016833333333333335,
      "logits/chosen": -0.39661893248558044,
      "logits/rejected": -0.4452442228794098,
      "logps/chosen": -146.19427490234375,
      "logps/rejected": -168.6130828857422,
      "loss": 0.1049,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -0.37566089630126953,
      "rewards/margins": 3.8498573303222656,
      "rewards/rejected": -4.225518226623535,
      "step": 50
    },
    {
      "epoch": 3.533333333333333,
      "grad_norm": 1.5647461414337158,
      "learning_rate": 0.00016,
      "logits/chosen": -0.9004085659980774,
      "logits/rejected": -0.9173569679260254,
      "logps/chosen": -164.2067413330078,
      "logps/rejected": -209.5588836669922,
      "loss": 0.0286,
      "rewards/accuracies": 0.9873417615890503,
      "rewards/chosen": -1.5753328800201416,
      "rewards/margins": 6.148852825164795,
      "rewards/rejected": -7.724185943603516,
      "step": 60
    },
    {
      "epoch": 4.118518518518519,
      "grad_norm": 0.05377896502614021,
      "learning_rate": 0.00015166666666666668,
      "logits/chosen": -1.1217529773712158,
      "logits/rejected": -1.1465003490447998,
      "logps/chosen": -170.9961395263672,
      "logps/rejected": -241.1328125,
      "loss": 0.0204,
      "rewards/accuracies": 0.9873417615890503,
      "rewards/chosen": -3.3284900188446045,
      "rewards/margins": 7.842370986938477,
      "rewards/rejected": -11.170860290527344,
      "step": 70
    },
    {
      "epoch": 4.711111111111111,
      "grad_norm": 0.05912935361266136,
      "learning_rate": 0.00014333333333333334,
      "logits/chosen": -1.3752106428146362,
      "logits/rejected": -1.3843052387237549,
      "logps/chosen": -200.71902465820312,
      "logps/rejected": -283.8058166503906,
      "loss": 0.0098,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -5.450322151184082,
      "rewards/margins": 9.867142677307129,
      "rewards/rejected": -15.317463874816895,
      "step": 80
    },
    {
      "epoch": 5.296296296296296,
      "grad_norm": 0.08618709444999695,
      "learning_rate": 0.00013500000000000003,
      "logits/chosen": -1.3256553411483765,
      "logits/rejected": -1.3671971559524536,
      "logps/chosen": -237.59788513183594,
      "logps/rejected": -331.6789245605469,
      "loss": 0.0103,
      "rewards/accuracies": 0.9873417615890503,
      "rewards/chosen": -9.06311321258545,
      "rewards/margins": 10.989459991455078,
      "rewards/rejected": -20.052576065063477,
      "step": 90
    },
    {
      "epoch": 5.888888888888889,
      "grad_norm": 0.0007632412016391754,
      "learning_rate": 0.00012666666666666666,
      "logits/chosen": -1.4357213973999023,
      "logits/rejected": -1.4478198289871216,
      "logps/chosen": -212.84860229492188,
      "logps/rejected": -312.7068786621094,
      "loss": 0.0176,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -6.595943450927734,
      "rewards/margins": 11.647254943847656,
      "rewards/rejected": -18.24319839477539,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}