{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.99581589958159,
  "eval_steps": 500,
  "global_step": 119,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.166666666666666e-08,
      "logits/chosen": -2.7454397678375244,
      "logits/rejected": -2.7223336696624756,
      "logps/chosen": -229.1898956298828,
      "logps/pi_response": -78.73236846923828,
      "logps/ref_response": -78.73236846923828,
      "logps/rejected": -192.8092041015625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -2.7830662727355957,
      "logits/rejected": -2.7350642681121826,
      "logps/chosen": -264.4526672363281,
      "logps/pi_response": -80.12222290039062,
      "logps/ref_response": -79.94595336914062,
      "logps/rejected": -162.01756286621094,
      "loss": 0.6918,
      "rewards/accuracies": 0.5486111044883728,
      "rewards/chosen": 0.002199761336669326,
      "rewards/margins": 0.0023648240603506565,
      "rewards/rejected": -0.00016506275278516114,
      "step": 10
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.931352528237397e-07,
      "logits/chosen": -2.7445321083068848,
      "logits/rejected": -2.714963436126709,
      "logps/chosen": -244.937255859375,
      "logps/pi_response": -78.12876892089844,
      "logps/ref_response": -72.18338775634766,
      "logps/rejected": -183.0408172607422,
      "loss": 0.6663,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": 0.03987501934170723,
      "rewards/margins": 0.07101120054721832,
      "rewards/rejected": -0.03113618493080139,
      "step": 20
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.658920803689553e-07,
      "logits/chosen": -2.640683650970459,
      "logits/rejected": -2.6123576164245605,
      "logps/chosen": -257.6654357910156,
      "logps/pi_response": -122.52134704589844,
      "logps/ref_response": -79.2972183227539,
      "logps/rejected": -204.9178009033203,
      "loss": 0.6232,
      "rewards/accuracies": 0.703125,
      "rewards/chosen": -0.09590627253055573,
      "rewards/margins": 0.1981358677148819,
      "rewards/rejected": -0.2940421402454376,
      "step": 30
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.201712553872657e-07,
      "logits/chosen": -2.5954911708831787,
      "logits/rejected": -2.592799663543701,
      "logps/chosen": -270.08135986328125,
      "logps/pi_response": -138.67002868652344,
      "logps/ref_response": -71.79570007324219,
      "logps/rejected": -237.4235382080078,
      "loss": 0.5903,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.2428603619337082,
      "rewards/margins": 0.3376618027687073,
      "rewards/rejected": -0.5805221796035767,
      "step": 40
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.598859066780754e-07,
      "logits/chosen": -2.6504998207092285,
      "logits/rejected": -2.6186861991882324,
      "logps/chosen": -259.39703369140625,
      "logps/pi_response": -165.39181518554688,
      "logps/ref_response": -67.79023742675781,
      "logps/rejected": -241.6773681640625,
      "loss": 0.553,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.42340484261512756,
      "rewards/margins": 0.43054357171058655,
      "rewards/rejected": -0.8539484143257141,
      "step": 50
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9019570347986706e-07,
      "logits/chosen": -2.664335250854492,
      "logits/rejected": -2.6367337703704834,
      "logps/chosen": -309.81878662109375,
      "logps/pi_response": -192.47970581054688,
      "logps/ref_response": -79.58308410644531,
      "logps/rejected": -292.54278564453125,
      "loss": 0.522,
      "rewards/accuracies": 0.7593749761581421,
      "rewards/chosen": -0.5549294352531433,
      "rewards/margins": 0.6205343008041382,
      "rewards/rejected": -1.1754637956619263,
      "step": 60
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1706525253979534e-07,
      "logits/chosen": -2.655745029449463,
      "logits/rejected": -2.6123197078704834,
      "logps/chosen": -315.2730407714844,
      "logps/pi_response": -216.87557983398438,
      "logps/ref_response": -83.69371032714844,
      "logps/rejected": -326.2763977050781,
      "loss": 0.4967,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -0.6697407364845276,
      "rewards/margins": 0.7002469301223755,
      "rewards/rejected": -1.3699876070022583,
      "step": 70
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4675360263490295e-07,
      "logits/chosen": -2.644761562347412,
      "logits/rejected": -2.6030020713806152,
      "logps/chosen": -328.7725830078125,
      "logps/pi_response": -233.19564819335938,
      "logps/ref_response": -71.0204849243164,
      "logps/rejected": -340.41845703125,
      "loss": 0.4833,
      "rewards/accuracies": 0.765625,
      "rewards/chosen": -0.8374919891357422,
      "rewards/margins": 0.9015256762504578,
      "rewards/rejected": -1.7390177249908447,
      "step": 80
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.527854855097224e-08,
      "logits/chosen": -2.6093177795410156,
      "logits/rejected": -2.5395705699920654,
      "logps/chosen": -379.70770263671875,
      "logps/pi_response": -282.5421447753906,
      "logps/ref_response": -77.24140930175781,
      "logps/rejected": -366.65423583984375,
      "loss": 0.4628,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -1.013519048690796,
      "rewards/margins": 1.0132733583450317,
      "rewards/rejected": -2.026792287826538,
      "step": 90
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.790158337517127e-08,
      "logits/chosen": -2.568589448928833,
      "logits/rejected": -2.4876933097839355,
      "logps/chosen": -330.4572448730469,
      "logps/pi_response": -253.181884765625,
      "logps/ref_response": -69.10757446289062,
      "logps/rejected": -371.88885498046875,
      "loss": 0.4734,
      "rewards/accuracies": 0.7906249761581421,
      "rewards/chosen": -0.9589544534683228,
      "rewards/margins": 0.9369667768478394,
      "rewards/rejected": -1.8959213495254517,
      "step": 100
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.677580722139671e-09,
      "logits/chosen": -2.6226189136505127,
      "logits/rejected": -2.55932354927063,
      "logps/chosen": -339.3543395996094,
      "logps/pi_response": -246.71542358398438,
      "logps/ref_response": -75.15101623535156,
      "logps/rejected": -361.17877197265625,
      "loss": 0.4478,
      "rewards/accuracies": 0.815625011920929,
      "rewards/chosen": -0.8720051050186157,
      "rewards/margins": 0.9231098294258118,
      "rewards/rejected": -1.7951151132583618,
      "step": 110
    },
    {
      "epoch": 1.0,
      "step": 119,
      "total_flos": 0.0,
      "train_loss": 0.5428742921652914,
      "train_runtime": 7148.4315,
      "train_samples_per_second": 4.276,
      "train_steps_per_second": 0.017
    }
  ],
  "logging_steps": 10,
  "max_steps": 119,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}