{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9968652037617555,
  "eval_steps": 500,
  "global_step": 159,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.125e-08,
      "logits/chosen": -1.9591341018676758,
      "logits/rejected": -2.0234761238098145,
      "logps/chosen": -395.7680969238281,
      "logps/rejected": -380.58642578125,
      "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": 3.1249999999999997e-07,
      "logits/chosen": -2.059805154800415,
      "logits/rejected": -1.9565424919128418,
      "logps/chosen": -263.6092224121094,
      "logps/rejected": -328.5196533203125,
      "loss": 0.6874,
      "rewards/accuracies": 0.4791666567325592,
      "rewards/chosen": -0.015080620534718037,
      "rewards/margins": 0.007691224571317434,
      "rewards/rejected": -0.022771844640374184,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.990353313429303e-07,
      "logits/chosen": -1.976946234703064,
      "logits/rejected": -1.9156849384307861,
      "logps/chosen": -268.0926818847656,
      "logps/rejected": -424.7269592285156,
      "loss": 0.647,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.2779475450515747,
      "rewards/margins": 0.2682579755783081,
      "rewards/rejected": -0.5462055206298828,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.882681251368548e-07,
      "logits/chosen": -1.9921376705169678,
      "logits/rejected": -1.928896188735962,
      "logps/chosen": -293.6935119628906,
      "logps/rejected": -397.9520263671875,
      "loss": 0.6707,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.22167165577411652,
      "rewards/margins": 0.3347076177597046,
      "rewards/rejected": -0.5563792586326599,
      "step": 30
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.6604720940421207e-07,
      "logits/chosen": -2.1060962677001953,
      "logits/rejected": -2.0056204795837402,
      "logps/chosen": -276.008544921875,
      "logps/rejected": -473.6175231933594,
      "loss": 0.6475,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.21760769188404083,
      "rewards/margins": 0.4528748393058777,
      "rewards/rejected": -0.6704825162887573,
      "step": 40
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3344075855595097e-07,
      "logits/chosen": -2.151175022125244,
      "logits/rejected": -1.9881855249404907,
      "logps/chosen": -283.6091613769531,
      "logps/rejected": -485.3069763183594,
      "loss": 0.6206,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.3180197775363922,
      "rewards/margins": 0.46803945302963257,
      "rewards/rejected": -0.7860592007637024,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.920161866827889e-07,
      "logits/chosen": -1.710820198059082,
      "logits/rejected": -1.5941283702850342,
      "logps/chosen": -286.6961364746094,
      "logps/rejected": -456.9854431152344,
      "loss": 0.6126,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.2723422348499298,
      "rewards/margins": 0.4845033288002014,
      "rewards/rejected": -0.7568455934524536,
      "step": 60
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.4376480090239047e-07,
      "logits/chosen": -1.174401044845581,
      "logits/rejected": -0.9936226010322571,
      "logps/chosen": -295.7551574707031,
      "logps/rejected": -473.75445556640625,
      "loss": 0.5755,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.42148083448410034,
      "rewards/margins": 0.6010136008262634,
      "rewards/rejected": -1.0224944353103638,
      "step": 70
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.910060778827554e-07,
      "logits/chosen": -1.2252156734466553,
      "logits/rejected": -0.9790323376655579,
      "logps/chosen": -318.60491943359375,
      "logps/rejected": -426.7520446777344,
      "loss": 0.5821,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.3635796904563904,
      "rewards/margins": 0.4388729929924011,
      "rewards/rejected": -0.8024527430534363,
      "step": 80
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.3627616503391812e-07,
      "logits/chosen": -1.1518232822418213,
      "logits/rejected": -0.7624595165252686,
      "logps/chosen": -298.54473876953125,
      "logps/rejected": -469.292236328125,
      "loss": 0.566,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.32996436953544617,
      "rewards/margins": 0.5770927667617798,
      "rewards/rejected": -0.9070571064949036,
      "step": 90
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8220596619089573e-07,
      "logits/chosen": -1.1035726070404053,
      "logits/rejected": -0.7693312764167786,
      "logps/chosen": -330.43011474609375,
      "logps/rejected": -488.72174072265625,
      "loss": 0.5851,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.515866219997406,
      "rewards/margins": 0.5535802245140076,
      "rewards/rejected": -1.0694465637207031,
      "step": 100
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3139467229135998e-07,
      "logits/chosen": -0.8986655473709106,
      "logits/rejected": -0.5390874147415161,
      "logps/chosen": -342.328857421875,
      "logps/rejected": -463.35302734375,
      "loss": 0.5799,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.6339472532272339,
      "rewards/margins": 0.45890823006629944,
      "rewards/rejected": -1.092855453491211,
      "step": 110
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.628481651367875e-08,
      "logits/chosen": -0.9509230852127075,
      "logits/rejected": -0.6183647513389587,
      "logps/chosen": -293.8681945800781,
      "logps/rejected": -508.28375244140625,
      "loss": 0.5524,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.47082701325416565,
      "rewards/margins": 0.6252096891403198,
      "rewards/rejected": -1.096036672592163,
      "step": 120
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.904486005914027e-08,
      "logits/chosen": -0.849618136882782,
      "logits/rejected": -0.5006222128868103,
      "logps/chosen": -314.4595031738281,
      "logps/rejected": -495.93621826171875,
      "loss": 0.5794,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.5189752578735352,
      "rewards/margins": 0.6143987774848938,
      "rewards/rejected": -1.1333739757537842,
      "step": 130
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.1464952759020856e-08,
      "logits/chosen": -0.6410492062568665,
      "logits/rejected": -0.3833368420600891,
      "logps/chosen": -327.976318359375,
      "logps/rejected": -475.7715759277344,
      "loss": 0.5506,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.6309124231338501,
      "rewards/margins": 0.559770941734314,
      "rewards/rejected": -1.190683364868164,
      "step": 140
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.8708793644441086e-09,
      "logits/chosen": -0.9400184750556946,
      "logits/rejected": -0.543064534664154,
      "logps/chosen": -335.626708984375,
      "logps/rejected": -510.7466735839844,
      "loss": 0.5794,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.5993916988372803,
      "rewards/margins": 0.5201231241226196,
      "rewards/rejected": -1.1195148229599,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 159,
      "total_flos": 0.0,
      "train_loss": 0.597387172890909,
      "train_runtime": 2649.9284,
      "train_samples_per_second": 7.69,
      "train_steps_per_second": 0.06
    }
  ],
  "logging_steps": 10,
  "max_steps": 159,
  "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
}