{
  "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": -2.58028244972229,
      "logits/rejected": -2.5264699459075928,
      "logps/chosen": -275.8560791015625,
      "logps/rejected": -243.97787475585938,
      "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": -2.599268913269043,
      "logits/rejected": -2.620181083679199,
      "logps/chosen": -229.47467041015625,
      "logps/rejected": -188.52099609375,
      "loss": 0.6925,
      "rewards/accuracies": 0.4756944477558136,
      "rewards/chosen": -0.005837044212967157,
      "rewards/margins": -1.28112105812761e-05,
      "rewards/rejected": -0.005824232939630747,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989935734988098e-06,
      "logits/chosen": -2.6293463706970215,
      "logits/rejected": -2.631697654724121,
      "logps/chosen": -233.75509643554688,
      "logps/rejected": -196.51675415039062,
      "loss": 0.6879,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.056370873004198074,
      "rewards/margins": 0.016503650695085526,
      "rewards/rejected": -0.0728745236992836,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.8776412907378845e-06,
      "logits/chosen": -2.6340599060058594,
      "logits/rejected": -2.6145126819610596,
      "logps/chosen": -244.6521453857422,
      "logps/rejected": -211.09170532226562,
      "loss": 0.6767,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.04168517515063286,
      "rewards/margins": 0.03752576559782028,
      "rewards/rejected": -0.07921093702316284,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": -2.5753064155578613,
      "logits/rejected": -2.546308994293213,
      "logps/chosen": -235.4735870361328,
      "logps/rejected": -216.3767852783203,
      "loss": 0.6724,
      "rewards/accuracies": 0.578125,
      "rewards/chosen": -0.20934326946735382,
      "rewards/margins": 0.032163362950086594,
      "rewards/rejected": -0.24150662124156952,
      "step": 40
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.3069871595684795e-06,
      "logits/chosen": -2.579021692276001,
      "logits/rejected": -2.5481619834899902,
      "logps/chosen": -238.4982452392578,
      "logps/rejected": -210.54537963867188,
      "loss": 0.6609,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.13248375058174133,
      "rewards/margins": 0.07907234132289886,
      "rewards/rejected": -0.211556077003479,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.8772424536302565e-06,
      "logits/chosen": -2.5088608264923096,
      "logits/rejected": -2.4528446197509766,
      "logps/chosen": -244.6417236328125,
      "logps/rejected": -219.36660766601562,
      "loss": 0.6588,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.10884495079517365,
      "rewards/margins": 0.0814167931675911,
      "rewards/rejected": -0.19026173651218414,
      "step": 60
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3784370602033572e-06,
      "logits/chosen": -2.4629616737365723,
      "logits/rejected": -2.438931941986084,
      "logps/chosen": -238.8616943359375,
      "logps/rejected": -205.7681121826172,
      "loss": 0.6535,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.17304827272891998,
      "rewards/margins": 0.10028227418661118,
      "rewards/rejected": -0.27333056926727295,
      "step": 70
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": -2.4710023403167725,
      "logits/rejected": -2.4444005489349365,
      "logps/chosen": -239.69384765625,
      "logps/rejected": -218.24093627929688,
      "loss": 0.6514,
      "rewards/accuracies": 0.5843750238418579,
      "rewards/chosen": -0.20169004797935486,
      "rewards/margins": 0.10011257976293564,
      "rewards/rejected": -0.3018026351928711,
      "step": 80
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2759017277414165e-06,
      "logits/chosen": -2.495241641998291,
      "logits/rejected": -2.4798271656036377,
      "logps/chosen": -267.980224609375,
      "logps/rejected": -255.43936157226562,
      "loss": 0.6519,
      "rewards/accuracies": 0.578125,
      "rewards/chosen": -0.25710397958755493,
      "rewards/margins": 0.0879761129617691,
      "rewards/rejected": -0.34508010745048523,
      "step": 90
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7274575140626318e-06,
      "logits/chosen": -2.4803011417388916,
      "logits/rejected": -2.4598233699798584,
      "logps/chosen": -257.54901123046875,
      "logps/rejected": -227.8817596435547,
      "loss": 0.6483,
      "rewards/accuracies": 0.640625,
      "rewards/chosen": -0.22189533710479736,
      "rewards/margins": 0.143667072057724,
      "rewards/rejected": -0.365562379360199,
      "step": 100
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -2.4257185459136963,
      "eval_logits/rejected": -2.339790105819702,
      "eval_logps/chosen": -308.5958557128906,
      "eval_logps/rejected": -300.188232421875,
      "eval_loss": 0.6988382339477539,
      "eval_rewards/accuracies": 0.4779999852180481,
      "eval_rewards/chosen": -0.10537321120500565,
      "eval_rewards/margins": 0.0014481162652373314,
      "eval_rewards/rejected": -0.10682134330272675,
      "eval_runtime": 383.862,
      "eval_samples_per_second": 5.21,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.217751806485235e-06,
      "logits/chosen": -2.4746994972229004,
      "logits/rejected": -2.4346601963043213,
      "logps/chosen": -272.1098327636719,
      "logps/rejected": -237.3122100830078,
      "loss": 0.6401,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.21971949934959412,
      "rewards/margins": 0.1589251309633255,
      "rewards/rejected": -0.3786446452140808,
      "step": 110
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -2.441807270050049,
      "logits/rejected": -2.439669609069824,
      "logps/chosen": -260.669189453125,
      "logps/rejected": -254.50344848632812,
      "loss": 0.6552,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.26529985666275024,
      "rewards/margins": 0.10932101309299469,
      "rewards/rejected": -0.3746209144592285,
      "step": 120
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1356686569674344e-07,
      "logits/chosen": -2.4392101764678955,
      "logits/rejected": -2.375783920288086,
      "logps/chosen": -241.207275390625,
      "logps/rejected": -224.5260772705078,
      "loss": 0.6479,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.24242636561393738,
      "rewards/margins": 0.09688840806484222,
      "rewards/rejected": -0.3393147885799408,
      "step": 130
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.59412823400657e-07,
      "logits/chosen": -2.3803234100341797,
      "logits/rejected": -2.3904504776000977,
      "logps/chosen": -245.4437255859375,
      "logps/rejected": -237.4853057861328,
      "loss": 0.6359,
      "rewards/accuracies": 0.628125011920929,
      "rewards/chosen": -0.213097482919693,
      "rewards/margins": 0.13748633861541748,
      "rewards/rejected": -0.35058385133743286,
      "step": 140
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.262559558016325e-08,
      "logits/chosen": -2.4386649131774902,
      "logits/rejected": -2.3991496562957764,
      "logps/chosen": -235.7268524169922,
      "logps/rejected": -208.8386688232422,
      "loss": 0.65,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.1908179074525833,
      "rewards/margins": 0.11969475448131561,
      "rewards/rejected": -0.3105126619338989,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 156,
      "total_flos": 0.0,
      "train_loss": 0.6591962063923861,
      "train_runtime": 7161.0752,
      "train_samples_per_second": 2.793,
      "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
}