{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 500,
  "global_step": 312,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.5625e-07,
      "logits/chosen": 0.37485986948013306,
      "logits/rejected": 0.6487500071525574,
      "logps/chosen": -1078.384765625,
      "logps/rejected": -1101.77490234375,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.5625e-06,
      "logits/chosen": 0.47351381182670593,
      "logits/rejected": 0.5273572206497192,
      "logps/chosen": -1056.42822265625,
      "logps/rejected": -1169.265625,
      "loss": 0.6932,
      "rewards/accuracies": 0.3958333432674408,
      "rewards/chosen": -0.00091694132424891,
      "rewards/margins": -0.00018613642896525562,
      "rewards/rejected": -0.0007308049243874848,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.125e-06,
      "logits/chosen": 0.42252635955810547,
      "logits/rejected": 0.49473732709884644,
      "logps/chosen": -1147.17236328125,
      "logps/rejected": -1265.768798828125,
      "loss": 0.6914,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": -0.022750383242964745,
      "rewards/margins": 0.0030057504773139954,
      "rewards/rejected": -0.02575613185763359,
      "step": 20
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.6875000000000004e-06,
      "logits/chosen": 0.5097017884254456,
      "logits/rejected": 0.5685985088348389,
      "logps/chosen": -1142.3890380859375,
      "logps/rejected": -1274.76171875,
      "loss": 0.6878,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.09474565088748932,
      "rewards/margins": 0.013504189439117908,
      "rewards/rejected": -0.10824984312057495,
      "step": 30
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989935734988098e-06,
      "logits/chosen": 0.6262896656990051,
      "logits/rejected": 0.5369861125946045,
      "logps/chosen": -1379.107177734375,
      "logps/rejected": -1572.9727783203125,
      "loss": 0.679,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.25016888976097107,
      "rewards/margins": 0.05141867324709892,
      "rewards/rejected": -0.3015875816345215,
      "step": 40
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.949188496058089e-06,
      "logits/chosen": 0.595242977142334,
      "logits/rejected": 0.6554633378982544,
      "logps/chosen": -1302.7606201171875,
      "logps/rejected": -1508.60107421875,
      "loss": 0.6637,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.35188037157058716,
      "rewards/margins": 0.07706869393587112,
      "rewards/rejected": -0.4289490282535553,
      "step": 50
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.8776412907378845e-06,
      "logits/chosen": 0.6328147053718567,
      "logits/rejected": 0.695043683052063,
      "logps/chosen": -1562.052978515625,
      "logps/rejected": -1646.878662109375,
      "loss": 0.6789,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": -0.4670296609401703,
      "rewards/margins": 0.025544878095388412,
      "rewards/rejected": -0.492574542760849,
      "step": 60
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.7761938666470405e-06,
      "logits/chosen": 0.5762341618537903,
      "logits/rejected": 0.6952670812606812,
      "logps/chosen": -1321.309814453125,
      "logps/rejected": -1596.348876953125,
      "loss": 0.6667,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.27966535091400146,
      "rewards/margins": 0.1273583322763443,
      "rewards/rejected": -0.40702366828918457,
      "step": 70
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": 0.5692261457443237,
      "logits/rejected": 0.8441296815872192,
      "logps/chosen": -1434.124755859375,
      "logps/rejected": -1714.9644775390625,
      "loss": 0.6653,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.38832995295524597,
      "rewards/margins": 0.12836144864559174,
      "rewards/rejected": -0.5166913866996765,
      "step": 80
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.4890613722044526e-06,
      "logits/chosen": 0.5933000445365906,
      "logits/rejected": 0.7363389730453491,
      "logps/chosen": -1363.330810546875,
      "logps/rejected": -1636.5830078125,
      "loss": 0.6614,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.36261868476867676,
      "rewards/margins": 0.11317511647939682,
      "rewards/rejected": -0.47579383850097656,
      "step": 90
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.3069871595684795e-06,
      "logits/chosen": 0.6000683903694153,
      "logits/rejected": 0.7250877618789673,
      "logps/chosen": -1501.1116943359375,
      "logps/rejected": -1688.3638916015625,
      "loss": 0.6729,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.42007994651794434,
      "rewards/margins": 0.07729745656251907,
      "rewards/rejected": -0.49737733602523804,
      "step": 100
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.102189034962561e-06,
      "logits/chosen": 0.6664993166923523,
      "logits/rejected": 0.8522599935531616,
      "logps/chosen": -1442.220703125,
      "logps/rejected": -1732.263671875,
      "loss": 0.6579,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.3821481168270111,
      "rewards/margins": 0.13758106529712677,
      "rewards/rejected": -0.5197292566299438,
      "step": 110
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.8772424536302565e-06,
      "logits/chosen": 0.7149074077606201,
      "logits/rejected": 0.7848154902458191,
      "logps/chosen": -1421.830322265625,
      "logps/rejected": -1694.971923828125,
      "loss": 0.6525,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.38273271918296814,
      "rewards/margins": 0.12498722970485687,
      "rewards/rejected": -0.5077199935913086,
      "step": 120
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.634976249348867e-06,
      "logits/chosen": 0.709136962890625,
      "logits/rejected": 0.7118825912475586,
      "logps/chosen": -1674.842041015625,
      "logps/rejected": -1917.5904541015625,
      "loss": 0.6612,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.5186460614204407,
      "rewards/margins": 0.10890078544616699,
      "rewards/rejected": -0.6275469064712524,
      "step": 130
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3784370602033572e-06,
      "logits/chosen": 0.6097074747085571,
      "logits/rejected": 0.8623794317245483,
      "logps/chosen": -1559.768798828125,
      "logps/rejected": -1758.3902587890625,
      "loss": 0.6673,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.446970134973526,
      "rewards/margins": 0.08759806305170059,
      "rewards/rejected": -0.534568190574646,
      "step": 140
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.1108510153447352e-06,
      "logits/chosen": 0.567806601524353,
      "logits/rejected": 0.8320453763008118,
      "logps/chosen": -1542.0999755859375,
      "logps/rejected": -1677.0787353515625,
      "loss": 0.6708,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.41464248299598694,
      "rewards/margins": 0.05952323600649834,
      "rewards/rejected": -0.47416573762893677,
      "step": 150
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": 0.7906177639961243,
      "logits/rejected": 0.7841562628746033,
      "logps/chosen": -1429.5552978515625,
      "logps/rejected": -1619.2171630859375,
      "loss": 0.6578,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.3786751627922058,
      "rewards/margins": 0.09149602800607681,
      "rewards/rejected": -0.47017115354537964,
      "step": 160
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.556095160739513e-06,
      "logits/chosen": 0.7081605195999146,
      "logits/rejected": 0.6871576905250549,
      "logps/chosen": -1445.623779296875,
      "logps/rejected": -1650.127197265625,
      "loss": 0.6577,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.41454702615737915,
      "rewards/margins": 0.10776009410619736,
      "rewards/rejected": -0.5223071575164795,
      "step": 170
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2759017277414165e-06,
      "logits/chosen": 0.7904581427574158,
      "logits/rejected": 0.7038453817367554,
      "logps/chosen": -1671.363525390625,
      "logps/rejected": -1868.4300537109375,
      "loss": 0.6579,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.5421901941299438,
      "rewards/margins": 0.10026909410953522,
      "rewards/rejected": -0.6424592733383179,
      "step": 180
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9985264605418185e-06,
      "logits/chosen": 0.6540366411209106,
      "logits/rejected": 0.7285584807395935,
      "logps/chosen": -1518.58935546875,
      "logps/rejected": -1726.029052734375,
      "loss": 0.652,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.4512515962123871,
      "rewards/margins": 0.09267839789390564,
      "rewards/rejected": -0.5439299941062927,
      "step": 190
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7274575140626318e-06,
      "logits/chosen": 0.879226565361023,
      "logits/rejected": 0.8555147051811218,
      "logps/chosen": -1627.2894287109375,
      "logps/rejected": -1928.3841552734375,
      "loss": 0.6602,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.5301727056503296,
      "rewards/margins": 0.161948561668396,
      "rewards/rejected": -0.6921212673187256,
      "step": 200
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.466103737583699e-06,
      "logits/chosen": 0.6861797571182251,
      "logits/rejected": 0.9016023874282837,
      "logps/chosen": -1539.524658203125,
      "logps/rejected": -1874.627685546875,
      "loss": 0.659,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.4866175651550293,
      "rewards/margins": 0.17361479997634888,
      "rewards/rejected": -0.6602323651313782,
      "step": 210
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.217751806485235e-06,
      "logits/chosen": 0.7002454996109009,
      "logits/rejected": 0.8032233119010925,
      "logps/chosen": -1692.8636474609375,
      "logps/rejected": -1974.6669921875,
      "loss": 0.6663,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.5267156958580017,
      "rewards/margins": 0.14205826818943024,
      "rewards/rejected": -0.6687740087509155,
      "step": 220
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.855248903979505e-07,
      "logits/chosen": 0.7965744733810425,
      "logits/rejected": 0.7885487079620361,
      "logps/chosen": -1604.53466796875,
      "logps/rejected": -1794.8245849609375,
      "loss": 0.6672,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.48480549454689026,
      "rewards/margins": 0.0899248868227005,
      "rewards/rejected": -0.574730396270752,
      "step": 230
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": 0.7397829294204712,
      "logits/rejected": 0.8248909711837769,
      "logps/chosen": -1530.4615478515625,
      "logps/rejected": -1826.6988525390625,
      "loss": 0.654,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.41139334440231323,
      "rewards/margins": 0.1389993578195572,
      "rewards/rejected": -0.550392746925354,
      "step": 240
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.808881491049723e-07,
      "logits/chosen": 0.6129003763198853,
      "logits/rejected": 0.8366864919662476,
      "logps/chosen": -1451.37890625,
      "logps/rejected": -1729.635986328125,
      "loss": 0.6728,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.40951260924339294,
      "rewards/margins": 0.1274668127298355,
      "rewards/rejected": -0.5369793772697449,
      "step": 250
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1356686569674344e-07,
      "logits/chosen": 0.7178203463554382,
      "logits/rejected": 0.8371836543083191,
      "logps/chosen": -1524.3497314453125,
      "logps/rejected": -1778.346923828125,
      "loss": 0.6495,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.4218766689300537,
      "rewards/margins": 0.12465916574001312,
      "rewards/rejected": -0.546535849571228,
      "step": 260
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.7248368952908055e-07,
      "logits/chosen": 0.6586390733718872,
      "logits/rejected": 0.8500372767448425,
      "logps/chosen": -1523.057373046875,
      "logps/rejected": -1753.3544921875,
      "loss": 0.6641,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.43091854453086853,
      "rewards/margins": 0.11140650510787964,
      "rewards/rejected": -0.5423250198364258,
      "step": 270
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.59412823400657e-07,
      "logits/chosen": 0.7076243162155151,
      "logits/rejected": 0.7846710681915283,
      "logps/chosen": -1433.0615234375,
      "logps/rejected": -1673.491455078125,
      "loss": 0.6713,
      "rewards/accuracies": 0.4937500059604645,
      "rewards/chosen": -0.39409512281417847,
      "rewards/margins": 0.11671394109725952,
      "rewards/rejected": -0.510809063911438,
      "step": 280
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.577619905828281e-08,
      "logits/chosen": 0.6640155911445618,
      "logits/rejected": 0.7537108659744263,
      "logps/chosen": -1413.724853515625,
      "logps/rejected": -1650.0406494140625,
      "loss": 0.6651,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.38546374440193176,
      "rewards/margins": 0.11493394523859024,
      "rewards/rejected": -0.5003976821899414,
      "step": 290
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.262559558016325e-08,
      "logits/chosen": 0.635712742805481,
      "logits/rejected": 0.8228713274002075,
      "logps/chosen": -1463.5419921875,
      "logps/rejected": -1637.050048828125,
      "loss": 0.6578,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.41111892461776733,
      "rewards/margins": 0.08340780436992645,
      "rewards/rejected": -0.4945267140865326,
      "step": 300
    },
    {
      "epoch": 0.99,
      "learning_rate": 6.294126437336734e-10,
      "logits/chosen": 0.6655277013778687,
      "logits/rejected": 0.759280800819397,
      "logps/chosen": -1619.28857421875,
      "logps/rejected": -1727.4703369140625,
      "loss": 0.6663,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.4628829061985016,
      "rewards/margins": 0.05492577701807022,
      "rewards/rejected": -0.517808735370636,
      "step": 310
    },
    {
      "epoch": 1.0,
      "step": 312,
      "total_flos": 0.0,
      "train_loss": 0.666122229435505,
      "train_runtime": 4198.8516,
      "train_samples_per_second": 4.763,
      "train_steps_per_second": 0.074
    }
  ],
  "logging_steps": 10,
  "max_steps": 312,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}