{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 100,
  "global_step": 156,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.125e-07,
      "logits/chosen": -2.749149799346924,
      "logits/rejected": -2.5982635021209717,
      "logps/chosen": -351.88592529296875,
      "logps/rejected": -339.1416931152344,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.125e-06,
      "logits/chosen": -2.634215831756592,
      "logits/rejected": -2.565009593963623,
      "logps/chosen": -264.03338623046875,
      "logps/rejected": -276.0550842285156,
      "loss": 0.6919,
      "rewards/accuracies": 0.5277777910232544,
      "rewards/chosen": -0.004038373474031687,
      "rewards/margins": 0.00274546816945076,
      "rewards/rejected": -0.006783841177821159,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989935734988098e-06,
      "logits/chosen": -2.6059346199035645,
      "logits/rejected": -2.5347509384155273,
      "logps/chosen": -289.2115173339844,
      "logps/rejected": -285.11651611328125,
      "loss": 0.6797,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.030138880014419556,
      "rewards/margins": 0.022945497184991837,
      "rewards/rejected": -0.05308438092470169,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.8776412907378845e-06,
      "logits/chosen": -2.549204111099243,
      "logits/rejected": -2.4668586254119873,
      "logps/chosen": -296.6357727050781,
      "logps/rejected": -286.880615234375,
      "loss": 0.6602,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.08948583900928497,
      "rewards/margins": 0.06929855048656464,
      "rewards/rejected": -0.1587844043970108,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": -2.607750177383423,
      "logits/rejected": -2.4750921726226807,
      "logps/chosen": -298.0829162597656,
      "logps/rejected": -306.82696533203125,
      "loss": 0.6514,
      "rewards/accuracies": 0.659375011920929,
      "rewards/chosen": -0.19320181012153625,
      "rewards/margins": 0.12534096837043762,
      "rewards/rejected": -0.3185427784919739,
      "step": 40
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.3069871595684795e-06,
      "logits/chosen": -2.520982265472412,
      "logits/rejected": -2.4599428176879883,
      "logps/chosen": -296.30560302734375,
      "logps/rejected": -325.0945129394531,
      "loss": 0.6459,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.12883062660694122,
      "rewards/margins": 0.13393087685108185,
      "rewards/rejected": -0.26276153326034546,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.8772424536302565e-06,
      "logits/chosen": -2.5419609546661377,
      "logits/rejected": -2.4526138305664062,
      "logps/chosen": -303.03753662109375,
      "logps/rejected": -309.76702880859375,
      "loss": 0.6405,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.13351231813430786,
      "rewards/margins": 0.12426233291625977,
      "rewards/rejected": -0.2577746510505676,
      "step": 60
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3784370602033572e-06,
      "logits/chosen": -2.4560115337371826,
      "logits/rejected": -2.3811163902282715,
      "logps/chosen": -344.1382141113281,
      "logps/rejected": -355.2807922363281,
      "loss": 0.6231,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.3589150905609131,
      "rewards/margins": 0.1809873878955841,
      "rewards/rejected": -0.5399025678634644,
      "step": 70
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": -2.4101650714874268,
      "logits/rejected": -2.2786593437194824,
      "logps/chosen": -342.35260009765625,
      "logps/rejected": -337.49755859375,
      "loss": 0.6147,
      "rewards/accuracies": 0.653124988079071,
      "rewards/chosen": -0.3759569525718689,
      "rewards/margins": 0.1979154795408249,
      "rewards/rejected": -0.573872447013855,
      "step": 80
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2759017277414165e-06,
      "logits/chosen": -2.2483856678009033,
      "logits/rejected": -2.258373260498047,
      "logps/chosen": -337.4295959472656,
      "logps/rejected": -386.83465576171875,
      "loss": 0.5889,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.4705071449279785,
      "rewards/margins": 0.3142710328102112,
      "rewards/rejected": -0.7847781181335449,
      "step": 90
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7274575140626318e-06,
      "logits/chosen": -2.268360137939453,
      "logits/rejected": -2.165742874145508,
      "logps/chosen": -334.42974853515625,
      "logps/rejected": -348.5438232421875,
      "loss": 0.5962,
      "rewards/accuracies": 0.684374988079071,
      "rewards/chosen": -0.475590318441391,
      "rewards/margins": 0.3093096613883972,
      "rewards/rejected": -0.7849000692367554,
      "step": 100
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -2.2236928939819336,
      "eval_logits/rejected": -2.0692262649536133,
      "eval_logps/chosen": -337.8192443847656,
      "eval_logps/rejected": -356.7066955566406,
      "eval_loss": 0.594050407409668,
      "eval_rewards/accuracies": 0.7020000219345093,
      "eval_rewards/chosen": -0.43988561630249023,
      "eval_rewards/margins": 0.31074145436286926,
      "eval_rewards/rejected": -0.7506271600723267,
      "eval_runtime": 384.0178,
      "eval_samples_per_second": 5.208,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.217751806485235e-06,
      "logits/chosen": -2.207293748855591,
      "logits/rejected": -1.979222059249878,
      "logps/chosen": -342.3262634277344,
      "logps/rejected": -362.3468933105469,
      "loss": 0.5798,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.526713490486145,
      "rewards/margins": 0.35861119627952576,
      "rewards/rejected": -0.8853246569633484,
      "step": 110
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -2.0455784797668457,
      "logits/rejected": -1.9951608180999756,
      "logps/chosen": -329.8724670410156,
      "logps/rejected": -370.48577880859375,
      "loss": 0.5914,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.56132572889328,
      "rewards/margins": 0.291836678981781,
      "rewards/rejected": -0.853162407875061,
      "step": 120
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1356686569674344e-07,
      "logits/chosen": -2.083890438079834,
      "logits/rejected": -1.9531793594360352,
      "logps/chosen": -333.07171630859375,
      "logps/rejected": -376.0746765136719,
      "loss": 0.5891,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -0.44948244094848633,
      "rewards/margins": 0.361866295337677,
      "rewards/rejected": -0.8113487362861633,
      "step": 130
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.59412823400657e-07,
      "logits/chosen": -2.076089382171631,
      "logits/rejected": -1.8212101459503174,
      "logps/chosen": -350.77374267578125,
      "logps/rejected": -360.770263671875,
      "loss": 0.5677,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.4118824005126953,
      "rewards/margins": 0.38872456550598145,
      "rewards/rejected": -0.8006070256233215,
      "step": 140
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.262559558016325e-08,
      "logits/chosen": -2.0566818714141846,
      "logits/rejected": -1.8309047222137451,
      "logps/chosen": -340.4973449707031,
      "logps/rejected": -353.11956787109375,
      "loss": 0.5863,
      "rewards/accuracies": 0.721875011920929,
      "rewards/chosen": -0.43235865235328674,
      "rewards/margins": 0.37298649549484253,
      "rewards/rejected": -0.8053451776504517,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 156,
      "total_flos": 0.0,
      "train_loss": 0.6200787784197391,
      "train_runtime": 7189.6884,
      "train_samples_per_second": 2.782,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}