{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9991836734693877,
  "eval_steps": 100,
  "global_step": 153,
  "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": -2.6383626461029053,
      "logits/rejected": -2.6590394973754883,
      "logps/chosen": -263.8804626464844,
      "logps/rejected": -342.5028381347656,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.125e-06,
      "logits/chosen": -2.692108392715454,
      "logits/rejected": -2.6769356727600098,
      "logps/chosen": -251.55091857910156,
      "logps/rejected": -279.13623046875,
      "loss": 0.693,
      "rewards/accuracies": 0.4444444477558136,
      "rewards/chosen": 0.006078645121306181,
      "rewards/margins": 0.001292299828492105,
      "rewards/rejected": 0.004786345642060041,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989490450759331e-06,
      "logits/chosen": -2.7136893272399902,
      "logits/rejected": -2.6634609699249268,
      "logps/chosen": -246.88778686523438,
      "logps/rejected": -270.17083740234375,
      "loss": 0.6909,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.009861673228442669,
      "rewards/margins": 0.002566133625805378,
      "rewards/rejected": 0.007295539136976004,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.872270441827174e-06,
      "logits/chosen": -2.7332160472869873,
      "logits/rejected": -2.6656548976898193,
      "logps/chosen": -276.451416015625,
      "logps/rejected": -291.919189453125,
      "loss": 0.6845,
      "rewards/accuracies": 0.578125,
      "rewards/chosen": -0.0339818000793457,
      "rewards/margins": 0.020258868113160133,
      "rewards/rejected": -0.05424066632986069,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.630851211353007e-06,
      "logits/chosen": -2.623366117477417,
      "logits/rejected": -2.5993411540985107,
      "logps/chosen": -253.49966430664062,
      "logps/rejected": -300.6990966796875,
      "loss": 0.6794,
      "rewards/accuracies": 0.590624988079071,
      "rewards/chosen": -0.047824542969465256,
      "rewards/margins": 0.032587505877017975,
      "rewards/rejected": -0.08041204512119293,
      "step": 40
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.277872161641682e-06,
      "logits/chosen": -2.695284366607666,
      "logits/rejected": -2.6641030311584473,
      "logps/chosen": -250.3493194580078,
      "logps/rejected": -272.2112731933594,
      "loss": 0.6766,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.006258129142224789,
      "rewards/margins": 0.02742757275700569,
      "rewards/rejected": -0.033685702830553055,
      "step": 50
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.831813362428005e-06,
      "logits/chosen": -2.6273930072784424,
      "logits/rejected": -2.5842039585113525,
      "logps/chosen": -262.25286865234375,
      "logps/rejected": -298.41510009765625,
      "loss": 0.667,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.15569987893104553,
      "rewards/margins": 0.06216084957122803,
      "rewards/rejected": -0.21786072850227356,
      "step": 60
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.3160280345958614e-06,
      "logits/chosen": -2.6327927112579346,
      "logits/rejected": -2.5868000984191895,
      "logps/chosen": -278.89703369140625,
      "logps/rejected": -314.3589172363281,
      "loss": 0.6649,
      "rewards/accuracies": 0.609375,
      "rewards/chosen": -0.1321047842502594,
      "rewards/margins": 0.05293331667780876,
      "rewards/rejected": -0.18503807485103607,
      "step": 70
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.757519902117886e-06,
      "logits/chosen": -2.6241769790649414,
      "logits/rejected": -2.61041522026062,
      "logps/chosen": -260.6992492675781,
      "logps/rejected": -272.4108581542969,
      "loss": 0.6678,
      "rewards/accuracies": 0.6031249761581421,
      "rewards/chosen": -0.16336451470851898,
      "rewards/margins": 0.07086249440908432,
      "rewards/rejected": -0.2342270165681839,
      "step": 80
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.185529423440807e-06,
      "logits/chosen": -2.6284656524658203,
      "logits/rejected": -2.6176791191101074,
      "logps/chosen": -282.1419372558594,
      "logps/rejected": -331.07354736328125,
      "loss": 0.6629,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.29410403966903687,
      "rewards/margins": 0.06631916761398315,
      "rewards/rejected": -0.3604232370853424,
      "step": 90
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.6300029195778454e-06,
      "logits/chosen": -2.631556749343872,
      "logits/rejected": -2.61484956741333,
      "logps/chosen": -280.27362060546875,
      "logps/rejected": -321.37884521484375,
      "loss": 0.6615,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.21054717898368835,
      "rewards/margins": 0.08309625834226608,
      "rewards/rejected": -0.2936434745788574,
      "step": 100
    },
    {
      "epoch": 0.65,
      "eval_logits/chosen": -2.6791629791259766,
      "eval_logits/rejected": -2.5814735889434814,
      "eval_logps/chosen": -310.7413024902344,
      "eval_logps/rejected": -294.6734619140625,
      "eval_loss": 0.6692679524421692,
      "eval_rewards/accuracies": 0.5519999861717224,
      "eval_rewards/chosen": -0.1884314864873886,
      "eval_rewards/margins": 0.05432567000389099,
      "eval_rewards/rejected": -0.2427571415901184,
      "eval_runtime": 395.8034,
      "eval_samples_per_second": 5.053,
      "eval_steps_per_second": 0.632,
      "step": 100
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.1200247470632394e-06,
      "logits/chosen": -2.6365230083465576,
      "logits/rejected": -2.636570453643799,
      "logps/chosen": -262.6191711425781,
      "logps/rejected": -310.3700256347656,
      "loss": 0.6646,
      "rewards/accuracies": 0.6343749761581421,
      "rewards/chosen": -0.22685377299785614,
      "rewards/margins": 0.07261677831411362,
      "rewards/rejected": -0.29947060346603394,
      "step": 110
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.822945986946386e-07,
      "logits/chosen": -2.6256182193756104,
      "logits/rejected": -2.5980124473571777,
      "logps/chosen": -246.5704803466797,
      "logps/rejected": -282.092529296875,
      "loss": 0.6621,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.20391853153705597,
      "rewards/margins": 0.08593297004699707,
      "rewards/rejected": -0.28985148668289185,
      "step": 120
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.397296523427807e-07,
      "logits/chosen": -2.584958553314209,
      "logits/rejected": -2.562863349914551,
      "logps/chosen": -262.963134765625,
      "logps/rejected": -306.772216796875,
      "loss": 0.6598,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.21135267615318298,
      "rewards/margins": 0.08709158003330231,
      "rewards/rejected": -0.2984442710876465,
      "step": 130
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.102647517397798e-07,
      "logits/chosen": -2.6206483840942383,
      "logits/rejected": -2.604473352432251,
      "logps/chosen": -272.1588439941406,
      "logps/rejected": -318.1951599121094,
      "loss": 0.6636,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.21283432841300964,
      "rewards/margins": 0.10337638854980469,
      "rewards/rejected": -0.3162107467651367,
      "step": 140
    },
    {
      "epoch": 0.98,
      "learning_rate": 5.9134352763748345e-09,
      "logits/chosen": -2.6338722705841064,
      "logits/rejected": -2.62086820602417,
      "logps/chosen": -296.335205078125,
      "logps/rejected": -331.77777099609375,
      "loss": 0.6594,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.23208002746105194,
      "rewards/margins": 0.08073471486568451,
      "rewards/rejected": -0.31281474232673645,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 153,
      "total_flos": 0.0,
      "train_loss": 0.6705375785141988,
      "train_runtime": 6538.0645,
      "train_samples_per_second": 2.998,
      "train_steps_per_second": 0.023
    }
  ],
  "logging_steps": 10,
  "max_steps": 153,
  "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
}