{
  "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": -2.387815237045288,
      "logits/rejected": -2.350198268890381,
      "logps/chosen": -242.59971618652344,
      "logps/rejected": -321.6684875488281,
      "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": -2.30749773979187,
      "logits/rejected": -2.2763028144836426,
      "logps/chosen": -245.24237060546875,
      "logps/rejected": -365.9026794433594,
      "loss": 0.6868,
      "rewards/accuracies": 0.4652777910232544,
      "rewards/chosen": -0.028837015852332115,
      "rewards/margins": 0.029528118669986725,
      "rewards/rejected": -0.05836513265967369,
      "step": 10
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.931352528237397e-07,
      "logits/chosen": -2.2129178047180176,
      "logits/rejected": -2.182603359222412,
      "logps/chosen": -280.6806945800781,
      "logps/rejected": -374.4796142578125,
      "loss": 0.6749,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.33705177903175354,
      "rewards/margins": 0.24108102917671204,
      "rewards/rejected": -0.5781327486038208,
      "step": 20
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.658920803689553e-07,
      "logits/chosen": -2.115509271621704,
      "logits/rejected": -2.1040070056915283,
      "logps/chosen": -292.1304626464844,
      "logps/rejected": -400.9358825683594,
      "loss": 0.6435,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.2881660759449005,
      "rewards/margins": 0.25809115171432495,
      "rewards/rejected": -0.5462571978569031,
      "step": 30
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.201712553872657e-07,
      "logits/chosen": -2.172727108001709,
      "logits/rejected": -2.1147096157073975,
      "logps/chosen": -268.44500732421875,
      "logps/rejected": -386.9407043457031,
      "loss": 0.6311,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.20380470156669617,
      "rewards/margins": 0.2536279559135437,
      "rewards/rejected": -0.45743265748023987,
      "step": 40
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.598859066780754e-07,
      "logits/chosen": -1.9897699356079102,
      "logits/rejected": -1.9906368255615234,
      "logps/chosen": -269.67657470703125,
      "logps/rejected": -444.35308837890625,
      "loss": 0.6442,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.2872712016105652,
      "rewards/margins": 0.3136231303215027,
      "rewards/rejected": -0.6008943319320679,
      "step": 50
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9019570347986706e-07,
      "logits/chosen": -2.0867788791656494,
      "logits/rejected": -2.029702663421631,
      "logps/chosen": -296.2376403808594,
      "logps/rejected": -376.4063415527344,
      "loss": 0.6373,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.32127830386161804,
      "rewards/margins": 0.1547112613916397,
      "rewards/rejected": -0.47598958015441895,
      "step": 60
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1706525253979534e-07,
      "logits/chosen": -2.0519118309020996,
      "logits/rejected": -2.0049712657928467,
      "logps/chosen": -324.88665771484375,
      "logps/rejected": -436.7549743652344,
      "loss": 0.6236,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.5307868719100952,
      "rewards/margins": 0.23191793262958527,
      "rewards/rejected": -0.7627049088478088,
      "step": 70
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4675360263490295e-07,
      "logits/chosen": -2.0299878120422363,
      "logits/rejected": -1.9803959131240845,
      "logps/chosen": -309.32440185546875,
      "logps/rejected": -413.36773681640625,
      "loss": 0.6451,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.4926892817020416,
      "rewards/margins": 0.29579848051071167,
      "rewards/rejected": -0.7884877920150757,
      "step": 80
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.527854855097224e-08,
      "logits/chosen": -2.030345916748047,
      "logits/rejected": -1.968796968460083,
      "logps/chosen": -303.46075439453125,
      "logps/rejected": -409.06549072265625,
      "loss": 0.6353,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.2820274829864502,
      "rewards/margins": 0.23102517426013947,
      "rewards/rejected": -0.5130526423454285,
      "step": 90
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.790158337517127e-08,
      "logits/chosen": -2.0697379112243652,
      "logits/rejected": -2.042279005050659,
      "logps/chosen": -274.68450927734375,
      "logps/rejected": -411.57318115234375,
      "loss": 0.6272,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.2681606113910675,
      "rewards/margins": 0.25202882289886475,
      "rewards/rejected": -0.5201894044876099,
      "step": 100
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.677580722139671e-09,
      "logits/chosen": -1.9937347173690796,
      "logits/rejected": -1.931985855102539,
      "logps/chosen": -270.1524963378906,
      "logps/rejected": -401.68719482421875,
      "loss": 0.6255,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.2766709327697754,
      "rewards/margins": 0.3142443597316742,
      "rewards/rejected": -0.590915322303772,
      "step": 110
    },
    {
      "epoch": 1.0,
      "step": 119,
      "total_flos": 0.0,
      "train_loss": 0.6447748657034225,
      "train_runtime": 1974.0915,
      "train_samples_per_second": 7.742,
      "train_steps_per_second": 0.06
    }
  ],
  "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
}