{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.6199095022624435,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.36199095022624433,
      "grad_norm": 4.351782321929932,
      "learning_rate": 9.259259259259259e-07,
      "logits/chosen": -2.3496108055114746,
      "logits/rejected": -2.354576349258423,
      "logps/chosen": -86.63139343261719,
      "logps/rejected": -76.90333557128906,
      "loss": 0.6923,
      "rewards/accuracies": 0.42500001192092896,
      "rewards/chosen": -0.00019822348258458078,
      "rewards/margins": -0.002337233629077673,
      "rewards/rejected": 0.00213901000097394,
      "step": 10
    },
    {
      "epoch": 0.7239819004524887,
      "grad_norm": 3.8085336685180664,
      "learning_rate": 1.8518518518518519e-06,
      "logits/chosen": -2.3482470512390137,
      "logits/rejected": -2.3420329093933105,
      "logps/chosen": -73.18113708496094,
      "logps/rejected": -72.52565002441406,
      "loss": 0.6943,
      "rewards/accuracies": 0.4000000059604645,
      "rewards/chosen": 0.00011640912998700514,
      "rewards/margins": -0.002153881825506687,
      "rewards/rejected": 0.002270291093736887,
      "step": 20
    },
    {
      "epoch": 1.085972850678733,
      "grad_norm": 5.329021453857422,
      "learning_rate": 2.7777777777777783e-06,
      "logits/chosen": -2.3689351081848145,
      "logits/rejected": -2.3885178565979004,
      "logps/chosen": -72.15180969238281,
      "logps/rejected": -74.70518493652344,
      "loss": 0.6924,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.0020837264601141214,
      "rewards/margins": 0.00603193324059248,
      "rewards/rejected": -0.0039482079446315765,
      "step": 30
    },
    {
      "epoch": 1.4479638009049773,
      "grad_norm": 3.671722888946533,
      "learning_rate": 3.7037037037037037e-06,
      "logits/chosen": -2.3091399669647217,
      "logits/rejected": -2.3273236751556396,
      "logps/chosen": -69.45216369628906,
      "logps/rejected": -260.0699157714844,
      "loss": 0.6899,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.0094170942902565,
      "rewards/margins": 0.14604035019874573,
      "rewards/rejected": -0.15545745193958282,
      "step": 40
    },
    {
      "epoch": 1.8099547511312217,
      "grad_norm": 4.172375679016113,
      "learning_rate": 4.62962962962963e-06,
      "logits/chosen": -2.31831693649292,
      "logits/rejected": -2.335618495941162,
      "logps/chosen": -72.82856750488281,
      "logps/rejected": -90.61243438720703,
      "loss": 0.6865,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.01846747286617756,
      "rewards/margins": 0.015672579407691956,
      "rewards/rejected": -0.034140050411224365,
      "step": 50
    },
    {
      "epoch": 2.171945701357466,
      "grad_norm": 4.386168479919434,
      "learning_rate": 4.998119881260576e-06,
      "logits/chosen": -2.3203041553497314,
      "logits/rejected": -2.3315658569335938,
      "logps/chosen": -67.68244934082031,
      "logps/rejected": -86.2356185913086,
      "loss": 0.6775,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.03993881493806839,
      "rewards/margins": 0.027381544932723045,
      "rewards/rejected": -0.06732036173343658,
      "step": 60
    },
    {
      "epoch": 2.5339366515837103,
      "grad_norm": 4.38440465927124,
      "learning_rate": 4.9866405060165044e-06,
      "logits/chosen": -2.3607311248779297,
      "logits/rejected": -2.3799984455108643,
      "logps/chosen": -72.8117904663086,
      "logps/rejected": -79.02535247802734,
      "loss": 0.6636,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.06544635444879532,
      "rewards/margins": 0.069185271859169,
      "rewards/rejected": -0.13463163375854492,
      "step": 70
    },
    {
      "epoch": 2.8959276018099547,
      "grad_norm": 4.333106517791748,
      "learning_rate": 4.964774158361991e-06,
      "logits/chosen": -2.392198324203491,
      "logits/rejected": -2.3854451179504395,
      "logps/chosen": -64.03775787353516,
      "logps/rejected": -71.01383972167969,
      "loss": 0.6473,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.09924488514661789,
      "rewards/margins": 0.1070803552865982,
      "rewards/rejected": -0.2063252478837967,
      "step": 80
    },
    {
      "epoch": 3.257918552036199,
      "grad_norm": 4.575820446014404,
      "learning_rate": 4.93261217644956e-06,
      "logits/chosen": -2.3316967487335205,
      "logits/rejected": -2.3399548530578613,
      "logps/chosen": -66.9199447631836,
      "logps/rejected": -67.97264099121094,
      "loss": 0.6292,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.1342274248600006,
      "rewards/margins": 0.1109505444765091,
      "rewards/rejected": -0.2451779842376709,
      "step": 90
    },
    {
      "epoch": 3.6199095022624435,
      "grad_norm": 4.666302680969238,
      "learning_rate": 4.8902889044347e-06,
      "logits/chosen": -2.3398795127868652,
      "logits/rejected": -2.3582355976104736,
      "logps/chosen": -76.05484771728516,
      "logps/rejected": -78.4997787475586,
      "loss": 0.6015,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.15528908371925354,
      "rewards/margins": 0.17627069354057312,
      "rewards/rejected": -0.33155977725982666,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 540,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.6988816913648845e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}