{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9971486761710794,
  "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": -2.8404829502105713,
      "logits/rejected": -2.7363457679748535,
      "logps/chosen": -137.61058044433594,
      "logps/rejected": -117.74949645996094,
      "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": -2.7154855728149414,
      "logits/rejected": -2.6928329467773438,
      "logps/chosen": -123.34764099121094,
      "logps/rejected": -126.27674865722656,
      "loss": 0.693,
      "rewards/accuracies": 0.4895833432674408,
      "rewards/chosen": 0.0009532907279208302,
      "rewards/margins": 0.00024687068071216345,
      "rewards/rejected": 0.0007064202218316495,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989490450759331e-06,
      "logits/chosen": -2.775507688522339,
      "logits/rejected": -2.7605175971984863,
      "logps/chosen": -131.35130310058594,
      "logps/rejected": -129.89755249023438,
      "loss": 0.6927,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": 0.012469060719013214,
      "rewards/margins": 0.00148476823233068,
      "rewards/rejected": 0.010984293185174465,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.872270441827174e-06,
      "logits/chosen": -2.7530579566955566,
      "logits/rejected": -2.703279733657837,
      "logps/chosen": -124.97566223144531,
      "logps/rejected": -123.0599136352539,
      "loss": 0.6914,
      "rewards/accuracies": 0.578125,
      "rewards/chosen": 0.04268375411629677,
      "rewards/margins": 0.004245065152645111,
      "rewards/rejected": 0.03843868523836136,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.630851211353007e-06,
      "logits/chosen": -2.7293448448181152,
      "logits/rejected": -2.70413875579834,
      "logps/chosen": -121.0544204711914,
      "logps/rejected": -125.32257080078125,
      "loss": 0.6899,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": 0.06691654026508331,
      "rewards/margins": 0.008759822696447372,
      "rewards/rejected": 0.05815672129392624,
      "step": 40
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.277872161641682e-06,
      "logits/chosen": -2.7005789279937744,
      "logits/rejected": -2.6852593421936035,
      "logps/chosen": -121.9286880493164,
      "logps/rejected": -122.9526596069336,
      "loss": 0.6889,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": 0.08492810279130936,
      "rewards/margins": 0.008367636241018772,
      "rewards/rejected": 0.07656045258045197,
      "step": 50
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.831813362428005e-06,
      "logits/chosen": -2.6874237060546875,
      "logits/rejected": -2.6661763191223145,
      "logps/chosen": -127.428955078125,
      "logps/rejected": -120.95658111572266,
      "loss": 0.6868,
      "rewards/accuracies": 0.5718749761581421,
      "rewards/chosen": 0.09333746135234833,
      "rewards/margins": 0.018141191452741623,
      "rewards/rejected": 0.075196273624897,
      "step": 60
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.3160280345958614e-06,
      "logits/chosen": -2.631645679473877,
      "logits/rejected": -2.592644214630127,
      "logps/chosen": -116.2786636352539,
      "logps/rejected": -119.904541015625,
      "loss": 0.6833,
      "rewards/accuracies": 0.621874988079071,
      "rewards/chosen": 0.068717822432518,
      "rewards/margins": 0.020535219460725784,
      "rewards/rejected": 0.04818259924650192,
      "step": 70
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.757519902117886e-06,
      "logits/chosen": -2.6794140338897705,
      "logits/rejected": -2.6262881755828857,
      "logps/chosen": -115.40583801269531,
      "logps/rejected": -118.37687683105469,
      "loss": 0.6826,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": 0.04097718372941017,
      "rewards/margins": 0.0270619448274374,
      "rewards/rejected": 0.01391523890197277,
      "step": 80
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.185529423440807e-06,
      "logits/chosen": -2.654372453689575,
      "logits/rejected": -2.6041693687438965,
      "logps/chosen": -125.3586654663086,
      "logps/rejected": -126.3048095703125,
      "loss": 0.6802,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": 0.07956322282552719,
      "rewards/margins": 0.03575668856501579,
      "rewards/rejected": 0.0438065305352211,
      "step": 90
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.6300029195778454e-06,
      "logits/chosen": -2.6299571990966797,
      "logits/rejected": -2.6014647483825684,
      "logps/chosen": -121.17118835449219,
      "logps/rejected": -126.56787109375,
      "loss": 0.6781,
      "rewards/accuracies": 0.590624988079071,
      "rewards/chosen": 0.06307043880224228,
      "rewards/margins": 0.035764340311288834,
      "rewards/rejected": 0.027306100353598595,
      "step": 100
    },
    {
      "epoch": 0.65,
      "eval_logits/chosen": -2.6276280879974365,
      "eval_logits/rejected": -2.5400655269622803,
      "eval_logps/chosen": -289.5703125,
      "eval_logps/rejected": -269.8819274902344,
      "eval_loss": 0.6668701171875,
      "eval_rewards/accuracies": 0.6439999938011169,
      "eval_rewards/chosen": -0.026262342929840088,
      "eval_rewards/margins": 0.059951718896627426,
      "eval_rewards/rejected": -0.08621405810117722,
      "eval_runtime": 381.9384,
      "eval_samples_per_second": 5.236,
      "eval_steps_per_second": 0.655,
      "step": 100
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.1200247470632394e-06,
      "logits/chosen": -2.640122652053833,
      "logits/rejected": -2.605874538421631,
      "logps/chosen": -128.22695922851562,
      "logps/rejected": -126.72957611083984,
      "loss": 0.6778,
      "rewards/accuracies": 0.565625011920929,
      "rewards/chosen": 0.03247489407658577,
      "rewards/margins": 0.034549370408058167,
      "rewards/rejected": -0.0020744793582707644,
      "step": 110
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.822945986946386e-07,
      "logits/chosen": -2.6430487632751465,
      "logits/rejected": -2.6212897300720215,
      "logps/chosen": -125.3406753540039,
      "logps/rejected": -127.47813415527344,
      "loss": 0.6773,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": 0.04354492202401161,
      "rewards/margins": 0.01878989301621914,
      "rewards/rejected": 0.02475503273308277,
      "step": 120
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.397296523427807e-07,
      "logits/chosen": -2.5850777626037598,
      "logits/rejected": -2.558905839920044,
      "logps/chosen": -126.96197509765625,
      "logps/rejected": -130.18173217773438,
      "loss": 0.676,
      "rewards/accuracies": 0.6343749761581421,
      "rewards/chosen": 0.026839906349778175,
      "rewards/margins": 0.039194256067276,
      "rewards/rejected": -0.012354351580142975,
      "step": 130
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.102647517397798e-07,
      "logits/chosen": -2.6094307899475098,
      "logits/rejected": -2.559872627258301,
      "logps/chosen": -123.62800598144531,
      "logps/rejected": -125.2538833618164,
      "loss": 0.6781,
      "rewards/accuracies": 0.590624988079071,
      "rewards/chosen": 0.02713879942893982,
      "rewards/margins": 0.04252254590392113,
      "rewards/rejected": -0.015383748337626457,
      "step": 140
    },
    {
      "epoch": 0.98,
      "learning_rate": 5.9134352763748345e-09,
      "logits/chosen": -2.6256508827209473,
      "logits/rejected": -2.61336088180542,
      "logps/chosen": -125.59254455566406,
      "logps/rejected": -130.33981323242188,
      "loss": 0.6745,
      "rewards/accuracies": 0.6031249761581421,
      "rewards/chosen": 0.032428476959466934,
      "rewards/margins": 0.05681389570236206,
      "rewards/rejected": -0.024385426193475723,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 153,
      "total_flos": 0.0,
      "train_loss": 0.6834432619069916,
      "train_runtime": 6282.2252,
      "train_samples_per_second": 3.125,
      "train_steps_per_second": 0.024
    }
  ],
  "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
}