{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 800000000,
  "global_step": 383,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.84375,
      "learning_rate": 1.282051282051282e-07,
      "logits/chosen": -3.3797154426574707,
      "logits/rejected": -3.440782070159912,
      "logps/chosen": -244.57943725585938,
      "logps/rejected": -168.14312744140625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/diff": -0.5416666865348816,
      "rewards/diff_abs": 0.5416666865348816,
      "rewards/rejected": 0.0,
      "rewards/student_margin": 0.0,
      "rewards/teacher_margin": 0.5416666865348816,
      "step": 1
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.875,
      "learning_rate": 1.282051282051282e-06,
      "logits/chosen": -3.357137680053711,
      "logits/rejected": -3.306201457977295,
      "logps/chosen": -323.62054443359375,
      "logps/rejected": -269.4796142578125,
      "loss": 0.6938,
      "rewards/accuracies": 0.48148149251937866,
      "rewards/chosen": -0.002036293037235737,
      "rewards/diff": -2.205963134765625,
      "rewards/diff_abs": 2.205963134765625,
      "rewards/rejected": 0.003695431165397167,
      "rewards/student_margin": -0.0057317232713103294,
      "rewards/teacher_margin": 2.2002317905426025,
      "step": 10
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.83984375,
      "learning_rate": 2.564102564102564e-06,
      "logits/chosen": -3.522096633911133,
      "logits/rejected": -3.589216709136963,
      "logps/chosen": -277.0390625,
      "logps/rejected": -192.60079956054688,
      "loss": 0.6932,
      "rewards/accuracies": 0.6333333253860474,
      "rewards/chosen": 0.004549483768641949,
      "rewards/diff": -2.0932600498199463,
      "rewards/diff_abs": 2.093954563140869,
      "rewards/rejected": -0.0068780360743403435,
      "rewards/student_margin": 0.011427519842982292,
      "rewards/teacher_margin": 2.1046876907348633,
      "step": 20
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.84765625,
      "learning_rate": 3.846153846153847e-06,
      "logits/chosen": -3.415924072265625,
      "logits/rejected": -3.5434460639953613,
      "logps/chosen": -301.97265625,
      "logps/rejected": -231.6956329345703,
      "loss": 0.6919,
      "rewards/accuracies": 0.36666664481163025,
      "rewards/chosen": -0.010005621239542961,
      "rewards/diff": -2.4664387702941895,
      "rewards/diff_abs": 2.4664387702941895,
      "rewards/rejected": 0.0007039735792204738,
      "rewards/student_margin": -0.010709594003856182,
      "rewards/teacher_margin": 2.4557292461395264,
      "step": 30
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.8203125,
      "learning_rate": 4.99989574668946e-06,
      "logits/chosen": -3.488649368286133,
      "logits/rejected": -3.546332597732544,
      "logps/chosen": -249.2800750732422,
      "logps/rejected": -179.78993225097656,
      "loss": 0.6901,
      "rewards/accuracies": 0.30000001192092896,
      "rewards/chosen": -0.007187096867710352,
      "rewards/diff": -2.8658556938171387,
      "rewards/diff_abs": 2.8658556938171387,
      "rewards/rejected": 0.010752001777291298,
      "rewards/student_margin": -0.017939096316695213,
      "rewards/teacher_margin": 2.847916841506958,
      "step": 40
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.8125,
      "learning_rate": 4.987395866955716e-06,
      "logits/chosen": -3.38765025138855,
      "logits/rejected": -3.5332159996032715,
      "logps/chosen": -331.1888427734375,
      "logps/rejected": -186.25814819335938,
      "loss": 0.6859,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": 0.018190687522292137,
      "rewards/diff": -2.0240583419799805,
      "rewards/diff_abs": 2.025913953781128,
      "rewards/rejected": -0.0025426391512155533,
      "rewards/student_margin": 0.02073333039879799,
      "rewards/teacher_margin": 2.0447916984558105,
      "step": 50
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.8515625,
      "learning_rate": 4.954164717534748e-06,
      "logits/chosen": -3.347484588623047,
      "logits/rejected": -3.380958080291748,
      "logps/chosen": -327.35113525390625,
      "logps/rejected": -350.645751953125,
      "loss": 0.6851,
      "rewards/accuracies": 0.5333333015441895,
      "rewards/chosen": 0.0018952175742015243,
      "rewards/diff": -1.218827486038208,
      "rewards/diff_abs": 1.2245365381240845,
      "rewards/rejected": 0.01603511907160282,
      "rewards/student_margin": -0.014139902777969837,
      "rewards/teacher_margin": 1.2046875953674316,
      "step": 60
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.8203125,
      "learning_rate": 4.900479264361017e-06,
      "logits/chosen": -3.400958299636841,
      "logits/rejected": -3.4466145038604736,
      "logps/chosen": -307.9151916503906,
      "logps/rejected": -278.41619873046875,
      "loss": 0.6795,
      "rewards/accuracies": 0.5999999642372131,
      "rewards/chosen": 0.03656778484582901,
      "rewards/diff": -1.3170411586761475,
      "rewards/diff_abs": 1.3170411586761475,
      "rewards/rejected": 0.030171453952789307,
      "rewards/student_margin": 0.0063963280990719795,
      "rewards/teacher_margin": 1.3234374523162842,
      "step": 70
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.8359375,
      "learning_rate": 4.826786950329646e-06,
      "logits/chosen": -3.5235488414764404,
      "logits/rejected": -3.580641269683838,
      "logps/chosen": -283.22747802734375,
      "logps/rejected": -180.744873046875,
      "loss": 0.6766,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": 0.0294550321996212,
      "rewards/diff": -1.3980296850204468,
      "rewards/diff_abs": 1.4047094583511353,
      "rewards/rejected": 0.005609627813100815,
      "rewards/student_margin": 0.023845406249165535,
      "rewards/teacher_margin": 1.421875,
      "step": 80
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.828125,
      "learning_rate": 4.733701966071226e-06,
      "logits/chosen": -3.4635212421417236,
      "logits/rejected": -3.515272617340088,
      "logps/chosen": -335.2662048339844,
      "logps/rejected": -171.05323791503906,
      "loss": 0.6755,
      "rewards/accuracies": 0.46666669845581055,
      "rewards/chosen": -0.00026724563213065267,
      "rewards/diff": -2.8980860710144043,
      "rewards/diff_abs": 2.8980860710144043,
      "rewards/rejected": 0.00875641219317913,
      "rewards/student_margin": -0.009023657068610191,
      "rewards/teacher_margin": 2.8890626430511475,
      "step": 90
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.82421875,
      "learning_rate": 4.622000130963015e-06,
      "logits/chosen": -3.5022895336151123,
      "logits/rejected": -3.5655486583709717,
      "logps/chosen": -305.0356750488281,
      "logps/rejected": -202.9454803466797,
      "loss": 0.6721,
      "rewards/accuracies": 0.5333333611488342,
      "rewards/chosen": 0.015491512604057789,
      "rewards/diff": -2.5059962272644043,
      "rewards/diff_abs": 2.5059962272644043,
      "rewards/rejected": -0.03372037410736084,
      "rewards/student_margin": 0.049211882054805756,
      "rewards/teacher_margin": 2.555208444595337,
      "step": 100
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.796875,
      "learning_rate": 4.492612427040864e-06,
      "logits/chosen": -3.5529093742370605,
      "logits/rejected": -3.6310908794403076,
      "logps/chosen": -277.04351806640625,
      "logps/rejected": -200.5059814453125,
      "loss": 0.6662,
      "rewards/accuracies": 0.7000000476837158,
      "rewards/chosen": 0.0211379025131464,
      "rewards/diff": -1.6043806076049805,
      "rewards/diff_abs": 1.6043806076049805,
      "rewards/rejected": -0.05364810302853584,
      "rewards/student_margin": 0.07478600740432739,
      "rewards/teacher_margin": 1.6791667938232422,
      "step": 110
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.8125,
      "learning_rate": 4.346617239703676e-06,
      "logits/chosen": -3.4835896492004395,
      "logits/rejected": -3.6081409454345703,
      "logps/chosen": -304.2657165527344,
      "logps/rejected": -239.1654052734375,
      "loss": 0.6626,
      "rewards/accuracies": 0.7333333492279053,
      "rewards/chosen": 0.052922703325748444,
      "rewards/diff": -1.7108633518218994,
      "rewards/diff_abs": 1.7417595386505127,
      "rewards/rejected": 0.003890041261911392,
      "rewards/student_margin": 0.04903266206383705,
      "rewards/teacher_margin": 1.7598956823349,
      "step": 120
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.83203125,
      "learning_rate": 4.185231369880461e-06,
      "logits/chosen": -3.2226901054382324,
      "logits/rejected": -3.4232337474823,
      "logps/chosen": -324.42236328125,
      "logps/rejected": -221.59872436523438,
      "loss": 0.6601,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.002616771962493658,
      "rewards/diff": -2.5576834678649902,
      "rewards/diff_abs": 2.5576834678649902,
      "rewards/rejected": -0.03972511366009712,
      "rewards/student_margin": 0.03710833936929703,
      "rewards/teacher_margin": 2.594791889190674,
      "step": 130
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.86328125,
      "learning_rate": 4.009799892569317e-06,
      "logits/chosen": -3.487971544265747,
      "logits/rejected": -3.4927310943603516,
      "logps/chosen": -294.6400451660156,
      "logps/rejected": -235.9231719970703,
      "loss": 0.6555,
      "rewards/accuracies": 0.8666666746139526,
      "rewards/chosen": 0.036725759506225586,
      "rewards/diff": -2.0255935192108154,
      "rewards/diff_abs": 2.0322105884552,
      "rewards/rejected": -0.10851386934518814,
      "rewards/student_margin": 0.14523963630199432,
      "rewards/teacher_margin": 2.1708333492279053,
      "step": 140
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.796875,
      "learning_rate": 3.8217849462726334e-06,
      "logits/chosen": -3.6184945106506348,
      "logits/rejected": -3.572862148284912,
      "logps/chosen": -246.6650390625,
      "logps/rejected": -221.8422088623047,
      "loss": 0.6532,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.019402923062443733,
      "rewards/diff": -1.9371099472045898,
      "rewards/diff_abs": 1.9371099472045898,
      "rewards/rejected": -0.05625145882368088,
      "rewards/student_margin": 0.03684854134917259,
      "rewards/teacher_margin": 1.9739586114883423,
      "step": 150
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.83203125,
      "learning_rate": 3.6227535467632873e-06,
      "logits/chosen": -3.5052731037139893,
      "logits/rejected": -3.6803977489471436,
      "logps/chosen": -441.88018798828125,
      "logps/rejected": -259.37799072265625,
      "loss": 0.647,
      "rewards/accuracies": 0.6333333253860474,
      "rewards/chosen": 0.04087870568037033,
      "rewards/diff": -1.968920350074768,
      "rewards/diff_abs": 1.9809789657592773,
      "rewards/rejected": -0.06728442758321762,
      "rewards/student_margin": 0.10816313326358795,
      "rewards/teacher_margin": 2.0770833492279053,
      "step": 160
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.8203125,
      "learning_rate": 3.4143645267483144e-06,
      "logits/chosen": -3.505716323852539,
      "logits/rejected": -3.5521907806396484,
      "logps/chosen": -318.2767333984375,
      "logps/rejected": -263.3311767578125,
      "loss": 0.6466,
      "rewards/accuracies": 0.7333332896232605,
      "rewards/chosen": -0.027162820100784302,
      "rewards/diff": -2.4665005207061768,
      "rewards/diff_abs": 2.5026025772094727,
      "rewards/rejected": -0.1044122725725174,
      "rewards/student_margin": 0.07724946737289429,
      "rewards/teacher_margin": 2.543750047683716,
      "step": 170
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.8125,
      "learning_rate": 3.1983547102818104e-06,
      "logits/chosen": -3.4710662364959717,
      "logits/rejected": -3.5504486560821533,
      "logps/chosen": -357.35101318359375,
      "logps/rejected": -294.84698486328125,
      "loss": 0.6401,
      "rewards/accuracies": 0.6333333253860474,
      "rewards/chosen": -0.05695001035928726,
      "rewards/diff": -1.7987518310546875,
      "rewards/diff_abs": 1.833298683166504,
      "rewards/rejected": -0.3196563422679901,
      "rewards/student_margin": 0.26270633935928345,
      "rewards/teacher_margin": 2.0614585876464844,
      "step": 180
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.81640625,
      "learning_rate": 2.9765244371567873e-06,
      "logits/chosen": -3.4952006340026855,
      "logits/rejected": -3.588458299636841,
      "logps/chosen": -280.5351257324219,
      "logps/rejected": -209.8900909423828,
      "loss": 0.638,
      "rewards/accuracies": 0.5666667222976685,
      "rewards/chosen": 0.002451175358146429,
      "rewards/diff": -2.4403529167175293,
      "rewards/diff_abs": 2.4539525508880615,
      "rewards/rejected": -0.18167506158351898,
      "rewards/student_margin": 0.18412622809410095,
      "rewards/teacher_margin": 2.624479055404663,
      "step": 190
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.83984375,
      "learning_rate": 2.7507225579233487e-06,
      "logits/chosen": -3.7310454845428467,
      "logits/rejected": -3.9183707237243652,
      "logps/chosen": -268.8802795410156,
      "logps/rejected": -197.14840698242188,
      "loss": 0.6356,
      "rewards/accuracies": 0.6999999284744263,
      "rewards/chosen": 0.019510583952069283,
      "rewards/diff": -2.137479305267334,
      "rewards/diff_abs": 2.137479305267334,
      "rewards/rejected": -0.06957288086414337,
      "rewards/student_margin": 0.0890834629535675,
      "rewards/teacher_margin": 2.2265625,
      "step": 200
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.875,
      "learning_rate": 2.522831024592615e-06,
      "logits/chosen": -3.594924211502075,
      "logits/rejected": -3.7799072265625,
      "logps/chosen": -307.20013427734375,
      "logps/rejected": -242.7568817138672,
      "loss": 0.6406,
      "rewards/accuracies": 0.6666666269302368,
      "rewards/chosen": -0.02844623290002346,
      "rewards/diff": -2.2553000450134277,
      "rewards/diff_abs": 2.259291172027588,
      "rewards/rejected": -0.13147947192192078,
      "rewards/student_margin": 0.10303322970867157,
      "rewards/teacher_margin": 2.3583333492279053,
      "step": 210
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.8359375,
      "learning_rate": 2.2947492054556075e-06,
      "logits/chosen": -3.5902042388916016,
      "logits/rejected": -3.788142681121826,
      "logps/chosen": -323.9666442871094,
      "logps/rejected": -212.97549438476562,
      "loss": 0.6286,
      "rewards/accuracies": 0.5666667222976685,
      "rewards/chosen": -0.08455105125904083,
      "rewards/diff": -1.5480363368988037,
      "rewards/diff_abs": 1.552331566810608,
      "rewards/rejected": -0.20005643367767334,
      "rewards/student_margin": 0.11550538241863251,
      "rewards/teacher_margin": 1.6635417938232422,
      "step": 220
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.8125,
      "learning_rate": 2.0683780547456666e-06,
      "logits/chosen": -3.5200886726379395,
      "logits/rejected": -3.7031588554382324,
      "logps/chosen": -314.7069396972656,
      "logps/rejected": -296.3868408203125,
      "loss": 0.6393,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": 0.02559809945523739,
      "rewards/diff": -1.6750246286392212,
      "rewards/diff_abs": 1.6955029964447021,
      "rewards/rejected": -0.23323163390159607,
      "rewards/student_margin": 0.2588297724723816,
      "rewards/teacher_margin": 1.933854341506958,
      "step": 230
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.8046875,
      "learning_rate": 1.845604269082787e-06,
      "logits/chosen": -3.597382068634033,
      "logits/rejected": -3.8518309593200684,
      "logps/chosen": -327.3302917480469,
      "logps/rejected": -229.868408203125,
      "loss": 0.6288,
      "rewards/accuracies": 0.5666666626930237,
      "rewards/chosen": -0.04607323184609413,
      "rewards/diff": -2.3076095581054688,
      "rewards/diff_abs": 2.3076095581054688,
      "rewards/rejected": -0.08013031631708145,
      "rewards/student_margin": 0.03405708074569702,
      "rewards/teacher_margin": 2.3416669368743896,
      "step": 240
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.81640625,
      "learning_rate": 1.628284562748429e-06,
      "logits/chosen": -3.5664756298065186,
      "logits/rejected": -3.988274097442627,
      "logps/chosen": -453.97540283203125,
      "logps/rejected": -193.4569854736328,
      "loss": 0.6248,
      "rewards/accuracies": 0.7333333492279053,
      "rewards/chosen": 0.06378475576639175,
      "rewards/diff": -2.4481723308563232,
      "rewards/diff_abs": 2.4774887561798096,
      "rewards/rejected": -0.18335548043251038,
      "rewards/student_margin": 0.24714021384716034,
      "rewards/teacher_margin": 2.695312738418579,
      "step": 250
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.82421875,
      "learning_rate": 1.4182301928489556e-06,
      "logits/chosen": -3.7244060039520264,
      "logits/rejected": -3.958916425704956,
      "logps/chosen": -319.8840026855469,
      "logps/rejected": -182.678955078125,
      "loss": 0.6236,
      "rewards/accuracies": 0.6333333253860474,
      "rewards/chosen": -0.004236244596540928,
      "rewards/diff": -2.6814849376678467,
      "rewards/diff_abs": 2.6881275177001953,
      "rewards/rejected": -0.15868888795375824,
      "rewards/student_margin": 0.1544526368379593,
      "rewards/teacher_margin": 2.835937738418579,
      "step": 260
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.7734375,
      "learning_rate": 1.2171918633431623e-06,
      "logits/chosen": -3.5800559520721436,
      "logits/rejected": -3.4622998237609863,
      "logps/chosen": -342.5908508300781,
      "logps/rejected": -346.6748352050781,
      "loss": 0.6325,
      "rewards/accuracies": 0.6999999284744263,
      "rewards/chosen": -0.052391983568668365,
      "rewards/diff": -1.884416937828064,
      "rewards/diff_abs": 1.9824146032333374,
      "rewards/rejected": -0.28151676058769226,
      "rewards/student_margin": 0.2291247844696045,
      "rewards/teacher_margin": 2.113541603088379,
      "step": 270
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.8046875,
      "learning_rate": 1.0268451337516774e-06,
      "logits/chosen": -3.7309062480926514,
      "logits/rejected": -3.933661699295044,
      "logps/chosen": -308.80218505859375,
      "logps/rejected": -169.55230712890625,
      "loss": 0.6301,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.07487048208713531,
      "rewards/diff": -2.276700735092163,
      "rewards/diff_abs": 2.2792751789093018,
      "rewards/rejected": -0.23931550979614258,
      "rewards/student_margin": 0.16444504261016846,
      "rewards/teacher_margin": 2.441145896911621,
      "step": 280
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.82421875,
      "learning_rate": 8.487764541597765e-07,
      "logits/chosen": -3.5284225940704346,
      "logits/rejected": -3.898432493209839,
      "logps/chosen": -268.7054748535156,
      "logps/rejected": -169.67959594726562,
      "loss": 0.6262,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.09594132751226425,
      "rewards/diff": -2.3169074058532715,
      "rewards/diff_abs": 2.3744282722473145,
      "rewards/rejected": -0.22174236178398132,
      "rewards/student_margin": 0.12580107152462006,
      "rewards/teacher_margin": 2.4427084922790527,
      "step": 290
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.81640625,
      "learning_rate": 6.844699429052377e-07,
      "logits/chosen": -3.428218364715576,
      "logits/rejected": -3.6805927753448486,
      "logps/chosen": -411.22412109375,
      "logps/rejected": -314.48919677734375,
      "loss": 0.6265,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.09454776346683502,
      "rewards/diff": -1.8355839252471924,
      "rewards/diff_abs": 1.851203203201294,
      "rewards/rejected": -0.2600056231021881,
      "rewards/student_margin": 0.16545787453651428,
      "rewards/teacher_margin": 2.001041889190674,
      "step": 300
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.83203125,
      "learning_rate": 5.352950171529928e-07,
      "logits/chosen": -3.615025281906128,
      "logits/rejected": -3.696945905685425,
      "logps/chosen": -245.61856079101562,
      "logps/rejected": -196.4246368408203,
      "loss": 0.6258,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -0.09576699882745743,
      "rewards/diff": -2.4770545959472656,
      "rewards/diff_abs": 2.491313934326172,
      "rewards/rejected": -0.31142088770866394,
      "rewards/student_margin": 0.2156539261341095,
      "rewards/teacher_margin": 2.6927084922790527,
      "step": 310
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.8203125,
      "learning_rate": 4.024949794498623e-07,
      "logits/chosen": -3.5860378742218018,
      "logits/rejected": -3.9259471893310547,
      "logps/chosen": -254.970458984375,
      "logps/rejected": -180.4410400390625,
      "loss": 0.6223,
      "rewards/accuracies": 0.7333333492279053,
      "rewards/chosen": -0.07610569894313812,
      "rewards/diff": -2.3922712802886963,
      "rewards/diff_abs": 2.509066104888916,
      "rewards/rejected": -0.33383452892303467,
      "rewards/student_margin": 0.2577288746833801,
      "rewards/teacher_margin": 2.6500000953674316,
      "step": 320
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.828125,
      "learning_rate": 2.8717665538507965e-07,
      "logits/chosen": -3.5933964252471924,
      "logits/rejected": -3.603943347930908,
      "logps/chosen": -280.0771789550781,
      "logps/rejected": -201.8335723876953,
      "loss": 0.6266,
      "rewards/accuracies": 0.6333333253860474,
      "rewards/chosen": -0.06104619428515434,
      "rewards/diff": -1.3486114740371704,
      "rewards/diff_abs": 1.5223140716552734,
      "rewards/rejected": -0.4061849117279053,
      "rewards/student_margin": 0.34513863921165466,
      "rewards/teacher_margin": 1.693750023841858,
      "step": 330
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.83984375,
      "learning_rate": 1.9030116872178317e-07,
      "logits/chosen": -3.7776057720184326,
      "logits/rejected": -3.7544455528259277,
      "logps/chosen": -294.9410705566406,
      "logps/rejected": -220.80538940429688,
      "loss": 0.6294,
      "rewards/accuracies": 0.6666666865348816,
      "rewards/chosen": -0.010373707860708237,
      "rewards/diff": -1.709843397140503,
      "rewards/diff_abs": 1.8078594207763672,
      "rewards/rejected": -0.30157214403152466,
      "rewards/student_margin": 0.2911984324455261,
      "rewards/teacher_margin": 2.001041889190674,
      "step": 340
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.8125,
      "learning_rate": 1.1267593088441886e-07,
      "logits/chosen": -3.633087158203125,
      "logits/rejected": -3.5318355560302734,
      "logps/chosen": -301.98785400390625,
      "logps/rejected": -271.0285949707031,
      "loss": 0.6263,
      "rewards/accuracies": 0.6333333253860474,
      "rewards/chosen": -0.04605107754468918,
      "rewards/diff": -1.565039873123169,
      "rewards/diff_abs": 1.6360604763031006,
      "rewards/rejected": -0.3101779520511627,
      "rewards/student_margin": 0.26412689685821533,
      "rewards/teacher_margin": 1.8291666507720947,
      "step": 350
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.83203125,
      "learning_rate": 5.494791156587686e-08,
      "logits/chosen": -3.7542724609375,
      "logits/rejected": -3.735842227935791,
      "logps/chosen": -233.6118621826172,
      "logps/rejected": -237.41378784179688,
      "loss": 0.6249,
      "rewards/accuracies": 0.6666666269302368,
      "rewards/chosen": -0.07444219291210175,
      "rewards/diff": -1.7430684566497803,
      "rewards/diff_abs": 1.7606357336044312,
      "rewards/rejected": -0.1824154257774353,
      "rewards/student_margin": 0.10797325521707535,
      "rewards/teacher_margin": 1.851041555404663,
      "step": 360
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.83984375,
      "learning_rate": 1.7598246540683483e-08,
      "logits/chosen": -3.840771436691284,
      "logits/rejected": -3.835493803024292,
      "logps/chosen": -247.64285278320312,
      "logps/rejected": -202.43807983398438,
      "loss": 0.6301,
      "rewards/accuracies": 0.7666666507720947,
      "rewards/chosen": -0.036110132932662964,
      "rewards/diff": -1.690749168395996,
      "rewards/diff_abs": 1.7279847860336304,
      "rewards/rejected": -0.28442350029945374,
      "rewards/student_margin": 0.24831333756446838,
      "rewards/teacher_margin": 1.9390627145767212,
      "step": 370
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.8125,
      "learning_rate": 9.382276255742729e-10,
      "logits/chosen": -3.547929286956787,
      "logits/rejected": -3.6237056255340576,
      "logps/chosen": -379.63031005859375,
      "logps/rejected": -313.1451721191406,
      "loss": 0.6277,
      "rewards/accuracies": 0.6999999284744263,
      "rewards/chosen": -0.0008453071350231767,
      "rewards/diff": -1.6570953130722046,
      "rewards/diff_abs": 1.7126853466033936,
      "rewards/rejected": -0.24843759834766388,
      "rewards/student_margin": 0.2475922852754593,
      "rewards/teacher_margin": 1.904687523841858,
      "step": 380
    },
    {
      "epoch": 1.0,
      "step": 383,
      "total_flos": 0.0,
      "train_loss": 0.6488963676806219,
      "train_runtime": 2900.3403,
      "train_samples_per_second": 50.684,
      "train_steps_per_second": 0.132
    }
  ],
  "logging_steps": 10,
  "max_steps": 383,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100000000000000000000000000000000,
  "total_flos": 0.0,
  "train_batch_size": 3,
  "trial_name": null,
  "trial_params": null
}