{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 100,
  "global_step": 195,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.5000000000000004e-07,
      "logits/chosen": -2.535799264907837,
      "logits/rejected": -2.5660998821258545,
      "logps/chosen": -270.7046813964844,
      "logps/rejected": -307.25311279296875,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.5e-06,
      "logits/chosen": -2.63254976272583,
      "logits/rejected": -2.5803234577178955,
      "logps/chosen": -320.9474792480469,
      "logps/rejected": -316.3360900878906,
      "loss": 0.6926,
      "rewards/accuracies": 0.4618055522441864,
      "rewards/chosen": -0.0115978317335248,
      "rewards/margins": 0.0027361076790839434,
      "rewards/rejected": -0.014333941973745823,
      "step": 10
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-06,
      "logits/chosen": -2.5984463691711426,
      "logits/rejected": -2.569882869720459,
      "logps/chosen": -307.0980529785156,
      "logps/rejected": -311.14959716796875,
      "loss": 0.6898,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": -0.03875315934419632,
      "rewards/margins": 0.007569065783172846,
      "rewards/rejected": -0.04632222652435303,
      "step": 20
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.959823971496575e-06,
      "logits/chosen": -2.5634453296661377,
      "logits/rejected": -2.4752097129821777,
      "logps/chosen": -322.8082580566406,
      "logps/rejected": -310.7356262207031,
      "loss": 0.6782,
      "rewards/accuracies": 0.6343749761581421,
      "rewards/chosen": -0.026555459946393967,
      "rewards/margins": 0.03691429644823074,
      "rewards/rejected": -0.06346975266933441,
      "step": 30
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.8405871765993435e-06,
      "logits/chosen": -2.55033278465271,
      "logits/rejected": -2.459357738494873,
      "logps/chosen": -333.2048034667969,
      "logps/rejected": -334.33538818359375,
      "loss": 0.6719,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.10182944685220718,
      "rewards/margins": 0.038650404661893845,
      "rewards/rejected": -0.14047986268997192,
      "step": 40
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": -2.530797243118286,
      "logits/rejected": -2.496924877166748,
      "logps/chosen": -313.50885009765625,
      "logps/rejected": -333.51397705078125,
      "loss": 0.6599,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.13014966249465942,
      "rewards/margins": 0.07707841694355011,
      "rewards/rejected": -0.20722806453704834,
      "step": 50
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.382678665009028e-06,
      "logits/chosen": -2.5104424953460693,
      "logits/rejected": -2.423055410385132,
      "logps/chosen": -337.88885498046875,
      "logps/rejected": -342.5820007324219,
      "loss": 0.6566,
      "rewards/accuracies": 0.609375,
      "rewards/chosen": -0.2016713172197342,
      "rewards/margins": 0.08087404072284698,
      "rewards/rejected": -0.2825453579425812,
      "step": 60
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.058724504646834e-06,
      "logits/chosen": -2.437530040740967,
      "logits/rejected": -2.411935329437256,
      "logps/chosen": -299.39825439453125,
      "logps/rejected": -334.86737060546875,
      "loss": 0.6486,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.1794266402721405,
      "rewards/margins": 0.12244921922683716,
      "rewards/rejected": -0.3018758296966553,
      "step": 70
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.684671656182497e-06,
      "logits/chosen": -2.494436740875244,
      "logits/rejected": -2.417968273162842,
      "logps/chosen": -309.20556640625,
      "logps/rejected": -324.0342102050781,
      "loss": 0.6481,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.22928757965564728,
      "rewards/margins": 0.13500627875328064,
      "rewards/rejected": -0.3642938435077667,
      "step": 80
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.272542485937369e-06,
      "logits/chosen": -2.3841819763183594,
      "logits/rejected": -2.356064558029175,
      "logps/chosen": -289.4866638183594,
      "logps/rejected": -311.481201171875,
      "loss": 0.6461,
      "rewards/accuracies": 0.590624988079071,
      "rewards/chosen": -0.2119632214307785,
      "rewards/margins": 0.12710149586200714,
      "rewards/rejected": -0.33906468749046326,
      "step": 90
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": -2.2960455417633057,
      "logits/rejected": -2.2413225173950195,
      "logps/chosen": -337.8830261230469,
      "logps/rejected": -365.25238037109375,
      "loss": 0.6429,
      "rewards/accuracies": 0.6156250238418579,
      "rewards/chosen": -0.3260457217693329,
      "rewards/margins": 0.16088134050369263,
      "rewards/rejected": -0.4869270920753479,
      "step": 100
    },
    {
      "epoch": 0.51,
      "eval_logits/chosen": -2.2674760818481445,
      "eval_logits/rejected": -2.134783983230591,
      "eval_logps/chosen": -357.25299072265625,
      "eval_logps/rejected": -365.74114990234375,
      "eval_loss": 0.6368065476417542,
      "eval_rewards/accuracies": 0.6380000114440918,
      "eval_rewards/chosen": -0.5885025858879089,
      "eval_rewards/margins": 0.17055951058864594,
      "eval_rewards/rejected": -0.7590621113777161,
      "eval_runtime": 383.7605,
      "eval_samples_per_second": 5.212,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.3878379241237136e-06,
      "logits/chosen": -2.160651445388794,
      "logits/rejected": -2.0833230018615723,
      "logps/chosen": -358.48834228515625,
      "logps/rejected": -367.35345458984375,
      "loss": 0.6391,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.5002601146697998,
      "rewards/margins": 0.17509731650352478,
      "rewards/rejected": -0.675357460975647,
      "step": 110
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9436976651092143e-06,
      "logits/chosen": -2.167994499206543,
      "logits/rejected": -2.064636707305908,
      "logps/chosen": -345.025390625,
      "logps/rejected": -360.7124938964844,
      "loss": 0.6299,
      "rewards/accuracies": 0.703125,
      "rewards/chosen": -0.29811879992485046,
      "rewards/margins": 0.1997375190258026,
      "rewards/rejected": -0.4978563189506531,
      "step": 120
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5174374208651913e-06,
      "logits/chosen": -1.9607616662979126,
      "logits/rejected": -1.7911456823349,
      "logps/chosen": -353.13446044921875,
      "logps/rejected": -370.7218017578125,
      "loss": 0.609,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.41445955634117126,
      "rewards/margins": 0.26371151208877563,
      "rewards/rejected": -0.6781710386276245,
      "step": 130
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.122757546369744e-06,
      "logits/chosen": -1.9263637065887451,
      "logits/rejected": -1.6748569011688232,
      "logps/chosen": -393.5077209472656,
      "logps/rejected": -390.3746643066406,
      "loss": 0.6161,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.4902144968509674,
      "rewards/margins": 0.21248309314250946,
      "rewards/rejected": -0.7026976346969604,
      "step": 140
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -1.7820335626602173,
      "logits/rejected": -1.6083834171295166,
      "logps/chosen": -354.44598388671875,
      "logps/rejected": -386.1634216308594,
      "loss": 0.6128,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.37745195627212524,
      "rewards/margins": 0.2488851547241211,
      "rewards/rejected": -0.6263370513916016,
      "step": 150
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.774575140626317e-07,
      "logits/chosen": -1.6063228845596313,
      "logits/rejected": -1.524730920791626,
      "logps/chosen": -350.2926330566406,
      "logps/rejected": -387.1871643066406,
      "loss": 0.6109,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.4679895043373108,
      "rewards/margins": 0.23797926306724548,
      "rewards/rejected": -0.7059687376022339,
      "step": 160
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.4757783024395244e-07,
      "logits/chosen": -1.4787060022354126,
      "logits/rejected": -1.4544020891189575,
      "logps/chosen": -342.1855773925781,
      "logps/rejected": -382.0643615722656,
      "loss": 0.62,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.46587371826171875,
      "rewards/margins": 0.2605627179145813,
      "rewards/rejected": -0.7264364361763,
      "step": 170
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.00928482603669e-08,
      "logits/chosen": -1.6203199625015259,
      "logits/rejected": -1.3692257404327393,
      "logps/chosen": -360.90191650390625,
      "logps/rejected": -369.82427978515625,
      "loss": 0.6112,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.4321191906929016,
      "rewards/margins": 0.2534908354282379,
      "rewards/rejected": -0.6856100559234619,
      "step": 180
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.006426501190233e-08,
      "logits/chosen": -1.5687239170074463,
      "logits/rejected": -1.3249285221099854,
      "logps/chosen": -354.10711669921875,
      "logps/rejected": -371.7186279296875,
      "loss": 0.6151,
      "rewards/accuracies": 0.653124988079071,
      "rewards/chosen": -0.44191044569015503,
      "rewards/margins": 0.21911220252513885,
      "rewards/rejected": -0.6610226035118103,
      "step": 190
    },
    {
      "epoch": 1.0,
      "step": 195,
      "total_flos": 0.0,
      "train_loss": 0.6412642754041231,
      "train_runtime": 8982.6903,
      "train_samples_per_second": 2.783,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 195,
  "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
}