{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9997382884061764,
  "eval_steps": 100,
  "global_step": 955,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5.208333333333334e-07,
      "logits/chosen": -2.980285167694092,
      "logits/rejected": -2.87275767326355,
      "logps/chosen": -313.4390563964844,
      "logps/rejected": -236.1754150390625,
      "loss": 0.6931,
      "pred_label": 0.0,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1,
      "use_label": 10.0
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.208333333333334e-06,
      "logits/chosen": -2.861464262008667,
      "logits/rejected": -2.907951593399048,
      "logps/chosen": -323.6517333984375,
      "logps/rejected": -284.9451904296875,
      "loss": 0.6921,
      "pred_label": 0.0,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": 0.0027037172112613916,
      "rewards/margins": 0.001292458618991077,
      "rewards/rejected": 0.0014112575445324183,
      "step": 10,
      "use_label": 90.0
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.0416666666666668e-05,
      "logits/chosen": -2.7527613639831543,
      "logits/rejected": -2.796025037765503,
      "logps/chosen": -236.6191864013672,
      "logps/rejected": -242.22232055664062,
      "loss": 0.6831,
      "pred_label": 0.0,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": 0.007748906500637531,
      "rewards/margins": 0.018490687012672424,
      "rewards/rejected": -0.010741781443357468,
      "step": 20,
      "use_label": 242.0
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.5625e-05,
      "logits/chosen": -2.8575313091278076,
      "logits/rejected": -2.829209804534912,
      "logps/chosen": -278.3554992675781,
      "logps/rejected": -252.61123657226562,
      "loss": 0.6574,
      "pred_label": 0.0,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": 0.026403894647955894,
      "rewards/margins": 0.09085250645875931,
      "rewards/rejected": -0.06444860994815826,
      "step": 30,
      "use_label": 402.0
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.0833333333333336e-05,
      "logits/chosen": -2.840946674346924,
      "logits/rejected": -2.8493659496307373,
      "logps/chosen": -281.32928466796875,
      "logps/rejected": -277.8607482910156,
      "loss": 0.6339,
      "pred_label": 0.0,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": 0.02641097828745842,
      "rewards/margins": 0.2079576551914215,
      "rewards/rejected": -0.1815466731786728,
      "step": 40,
      "use_label": 562.0
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.604166666666667e-05,
      "logits/chosen": -2.854478120803833,
      "logits/rejected": -2.839433193206787,
      "logps/chosen": -266.80517578125,
      "logps/rejected": -261.9176025390625,
      "loss": 0.5914,
      "pred_label": 1.2999999523162842,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.06968289613723755,
      "rewards/margins": 0.33043327927589417,
      "rewards/rejected": -0.4001162648200989,
      "step": 50,
      "use_label": 720.7000122070312
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.125e-05,
      "logits/chosen": -2.818674087524414,
      "logits/rejected": -2.8088903427124023,
      "logps/chosen": -300.7886962890625,
      "logps/rejected": -290.3478698730469,
      "loss": 0.6053,
      "pred_label": 5.875,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.03734510391950607,
      "rewards/margins": 0.38690200448036194,
      "rewards/rejected": -0.4242470860481262,
      "step": 60,
      "use_label": 876.125
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.6458333333333336e-05,
      "logits/chosen": -2.8236336708068848,
      "logits/rejected": -2.818040132522583,
      "logps/chosen": -294.469970703125,
      "logps/rejected": -278.6357421875,
      "loss": 0.535,
      "pred_label": 13.5,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.015645451843738556,
      "rewards/margins": 0.5513776540756226,
      "rewards/rejected": -0.5670231580734253,
      "step": 70,
      "use_label": 1028.5
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.166666666666667e-05,
      "logits/chosen": -2.770120143890381,
      "logits/rejected": -2.756917953491211,
      "logps/chosen": -306.6744384765625,
      "logps/rejected": -288.44793701171875,
      "loss": 0.4709,
      "pred_label": 36.79999923706055,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -0.13690751791000366,
      "rewards/margins": 0.8066266179084778,
      "rewards/rejected": -0.9435340762138367,
      "step": 80,
      "use_label": 1165.199951171875
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.6875e-05,
      "logits/chosen": -2.698793411254883,
      "logits/rejected": -2.75233793258667,
      "logps/chosen": -295.5422668457031,
      "logps/rejected": -274.40032958984375,
      "loss": 0.5055,
      "pred_label": 72.92500305175781,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": 0.02409188821911812,
      "rewards/margins": 0.8366730809211731,
      "rewards/rejected": -0.8125811815261841,
      "step": 90,
      "use_label": 1289.074951171875
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.976717112922003e-05,
      "logits/chosen": -2.719881057739258,
      "logits/rejected": -2.7187676429748535,
      "logps/chosen": -276.2088317871094,
      "logps/rejected": -304.5025329589844,
      "loss": 0.5147,
      "pred_label": 106.19999694824219,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.20771054923534393,
      "rewards/margins": 0.8888995051383972,
      "rewards/rejected": -1.096610188484192,
      "step": 100,
      "use_label": 1415.800048828125
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.918509895227008e-05,
      "logits/chosen": -2.7200231552124023,
      "logits/rejected": -2.6798348426818848,
      "logps/chosen": -266.4808044433594,
      "logps/rejected": -259.0126037597656,
      "loss": 0.5078,
      "pred_label": 140.47500610351562,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.18062053620815277,
      "rewards/margins": 0.8032993078231812,
      "rewards/rejected": -0.9839197993278503,
      "step": 110,
      "use_label": 1541.5250244140625
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.860302677532014e-05,
      "logits/chosen": -2.7260735034942627,
      "logits/rejected": -2.729832172393799,
      "logps/chosen": -280.414306640625,
      "logps/rejected": -256.8193664550781,
      "loss": 0.5022,
      "pred_label": 172.5,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.09014640003442764,
      "rewards/margins": 0.8234178423881531,
      "rewards/rejected": -0.9135642051696777,
      "step": 120,
      "use_label": 1669.5
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.80209545983702e-05,
      "logits/chosen": -2.7872612476348877,
      "logits/rejected": -2.748222589492798,
      "logps/chosen": -308.61865234375,
      "logps/rejected": -281.79461669921875,
      "loss": 0.5032,
      "pred_label": 206.75,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.1351226270198822,
      "rewards/margins": 0.7332055568695068,
      "rewards/rejected": -0.8683282136917114,
      "step": 130,
      "use_label": 1795.25
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.743888242142026e-05,
      "logits/chosen": -2.794041872024536,
      "logits/rejected": -2.7909293174743652,
      "logps/chosen": -260.35333251953125,
      "logps/rejected": -247.11325073242188,
      "loss": 0.4325,
      "pred_label": 246.6999969482422,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.22465598583221436,
      "rewards/margins": 0.984777569770813,
      "rewards/rejected": -1.209433674812317,
      "step": 140,
      "use_label": 1915.300048828125
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.685681024447032e-05,
      "logits/chosen": -2.705472946166992,
      "logits/rejected": -2.7284083366394043,
      "logps/chosen": -300.3094787597656,
      "logps/rejected": -294.97125244140625,
      "loss": 0.4364,
      "pred_label": 291.7250061035156,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.5423473119735718,
      "rewards/margins": 0.7800337076187134,
      "rewards/rejected": -1.3223809003829956,
      "step": 150,
      "use_label": 2030.2750244140625
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.6274738067520374e-05,
      "logits/chosen": -2.743900775909424,
      "logits/rejected": -2.688366174697876,
      "logps/chosen": -263.2331237792969,
      "logps/rejected": -281.53582763671875,
      "loss": 0.4222,
      "pred_label": 345.32501220703125,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.0851084440946579,
      "rewards/margins": 1.1642277240753174,
      "rewards/rejected": -1.2493362426757812,
      "step": 160,
      "use_label": 2136.675048828125
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.5692665890570435e-05,
      "logits/chosen": -2.7768936157226562,
      "logits/rejected": -2.779693126678467,
      "logps/chosen": -297.79376220703125,
      "logps/rejected": -261.3406982421875,
      "loss": 0.4307,
      "pred_label": 394.0,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.13176263868808746,
      "rewards/margins": 1.0377919673919678,
      "rewards/rejected": -1.169554591178894,
      "step": 170,
      "use_label": 2248.0
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.511059371362049e-05,
      "logits/chosen": -2.6910805702209473,
      "logits/rejected": -2.6632163524627686,
      "logps/chosen": -259.5965270996094,
      "logps/rejected": -289.4101257324219,
      "loss": 0.4077,
      "pred_label": 464.125,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.3667333722114563,
      "rewards/margins": 1.5060479640960693,
      "rewards/rejected": -1.8727811574935913,
      "step": 180,
      "use_label": 2337.875
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.452852153667055e-05,
      "logits/chosen": -2.7862701416015625,
      "logits/rejected": -2.747239589691162,
      "logps/chosen": -308.29376220703125,
      "logps/rejected": -280.5757751464844,
      "loss": 0.3987,
      "pred_label": 532.6500244140625,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": -0.10377927869558334,
      "rewards/margins": 1.8495817184448242,
      "rewards/rejected": -1.953360915184021,
      "step": 190,
      "use_label": 2429.35009765625
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.394644935972061e-05,
      "logits/chosen": -2.787761688232422,
      "logits/rejected": -2.7670867443084717,
      "logps/chosen": -267.0201110839844,
      "logps/rejected": -287.81341552734375,
      "loss": 0.4013,
      "pred_label": 604.4000244140625,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.5691341757774353,
      "rewards/margins": 1.3881704807281494,
      "rewards/rejected": -1.95730459690094,
      "step": 200,
      "use_label": 2517.60009765625
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.336437718277067e-05,
      "logits/chosen": -2.7742085456848145,
      "logits/rejected": -2.794970989227295,
      "logps/chosen": -306.9640808105469,
      "logps/rejected": -271.60906982421875,
      "loss": 0.4304,
      "pred_label": 661.4000244140625,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -0.5716164708137512,
      "rewards/margins": 1.1703290939331055,
      "rewards/rejected": -1.7419456243515015,
      "step": 210,
      "use_label": 2620.60009765625
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.278230500582072e-05,
      "logits/chosen": -2.7252678871154785,
      "logits/rejected": -2.740291118621826,
      "logps/chosen": -280.5372009277344,
      "logps/rejected": -290.718994140625,
      "loss": 0.3818,
      "pred_label": 722.1749877929688,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.5545529723167419,
      "rewards/margins": 1.4516589641571045,
      "rewards/rejected": -2.006211996078491,
      "step": 220,
      "use_label": 2719.824951171875
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.220023282887078e-05,
      "logits/chosen": -2.7117373943328857,
      "logits/rejected": -2.7296900749206543,
      "logps/chosen": -316.2425231933594,
      "logps/rejected": -287.38421630859375,
      "loss": 0.3674,
      "pred_label": 794.0499877929688,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.8069397211074829,
      "rewards/margins": 1.5205641984939575,
      "rewards/rejected": -2.3275039196014404,
      "step": 230,
      "use_label": 2807.949951171875
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.161816065192084e-05,
      "logits/chosen": -2.7252871990203857,
      "logits/rejected": -2.7065608501434326,
      "logps/chosen": -281.2114562988281,
      "logps/rejected": -245.62576293945312,
      "loss": 0.4159,
      "pred_label": 862.6500244140625,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.6634035706520081,
      "rewards/margins": 1.5971952676773071,
      "rewards/rejected": -2.260598659515381,
      "step": 240,
      "use_label": 2899.35009765625
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.10360884749709e-05,
      "logits/chosen": -2.7199912071228027,
      "logits/rejected": -2.709681987762451,
      "logps/chosen": -296.9573974609375,
      "logps/rejected": -291.0780334472656,
      "loss": 0.4335,
      "pred_label": 924.7000122070312,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.63172847032547,
      "rewards/margins": 1.5341920852661133,
      "rewards/rejected": -2.1659207344055176,
      "step": 250,
      "use_label": 2997.300048828125
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.045401629802096e-05,
      "logits/chosen": -2.7235524654388428,
      "logits/rejected": -2.7239880561828613,
      "logps/chosen": -256.81951904296875,
      "logps/rejected": -270.71209716796875,
      "loss": 0.4011,
      "pred_label": 982.9749755859375,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.8210113644599915,
      "rewards/margins": 1.2212928533554077,
      "rewards/rejected": -2.042304277420044,
      "step": 260,
      "use_label": 3099.02490234375
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.9871944121071014e-05,
      "logits/chosen": -2.7312769889831543,
      "logits/rejected": -2.739067554473877,
      "logps/chosen": -298.7057800292969,
      "logps/rejected": -300.0126953125,
      "loss": 0.4119,
      "pred_label": 1049.5,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -0.6834469437599182,
      "rewards/margins": 1.8426272869110107,
      "rewards/rejected": -2.5260744094848633,
      "step": 270,
      "use_label": 3192.5
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.928987194412107e-05,
      "logits/chosen": -2.807619333267212,
      "logits/rejected": -2.8070969581604004,
      "logps/chosen": -298.39654541015625,
      "logps/rejected": -282.5412902832031,
      "loss": 0.401,
      "pred_label": 1114.199951171875,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -0.44325321912765503,
      "rewards/margins": 1.5879560708999634,
      "rewards/rejected": -2.0312094688415527,
      "step": 280,
      "use_label": 3287.800048828125
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.870779976717113e-05,
      "logits/chosen": -2.7824301719665527,
      "logits/rejected": -2.762150764465332,
      "logps/chosen": -300.687255859375,
      "logps/rejected": -250.01663208007812,
      "loss": 0.3923,
      "pred_label": 1174.4749755859375,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.7764869928359985,
      "rewards/margins": 1.2035253047943115,
      "rewards/rejected": -1.9800125360488892,
      "step": 290,
      "use_label": 3387.52490234375
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.812572759022119e-05,
      "logits/chosen": -2.769491195678711,
      "logits/rejected": -2.732734203338623,
      "logps/chosen": -271.75091552734375,
      "logps/rejected": -263.69464111328125,
      "loss": 0.4009,
      "pred_label": 1240.0250244140625,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.8072065114974976,
      "rewards/margins": 1.5778390169143677,
      "rewards/rejected": -2.3850455284118652,
      "step": 300,
      "use_label": 3481.97509765625
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.7543655413271246e-05,
      "logits/chosen": -2.7556052207946777,
      "logits/rejected": -2.6895577907562256,
      "logps/chosen": -273.4451904296875,
      "logps/rejected": -317.7626953125,
      "loss": 0.3834,
      "pred_label": 1312.050048828125,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -0.6513568162918091,
      "rewards/margins": 2.270569324493408,
      "rewards/rejected": -2.921926259994507,
      "step": 310,
      "use_label": 3569.949951171875
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.696158323632131e-05,
      "logits/chosen": -2.73972749710083,
      "logits/rejected": -2.7072513103485107,
      "logps/chosen": -303.25665283203125,
      "logps/rejected": -305.5859069824219,
      "loss": 0.3917,
      "pred_label": 1381.9000244140625,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -0.7276700735092163,
      "rewards/margins": 1.7501161098480225,
      "rewards/rejected": -2.4777863025665283,
      "step": 320,
      "use_label": 3660.10009765625
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.637951105937136e-05,
      "logits/chosen": -2.793794631958008,
      "logits/rejected": -2.7443809509277344,
      "logps/chosen": -277.2012634277344,
      "logps/rejected": -254.2499542236328,
      "loss": 0.4169,
      "pred_label": 1455.949951171875,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.6252767443656921,
      "rewards/margins": 1.3715183734893799,
      "rewards/rejected": -1.9967950582504272,
      "step": 330,
      "use_label": 3746.050048828125
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.579743888242142e-05,
      "logits/chosen": -2.747743606567383,
      "logits/rejected": -2.763869524002075,
      "logps/chosen": -294.94415283203125,
      "logps/rejected": -256.500244140625,
      "loss": 0.4603,
      "pred_label": 1514.300048828125,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.5764142870903015,
      "rewards/margins": 0.9885671734809875,
      "rewards/rejected": -1.564981460571289,
      "step": 340,
      "use_label": 3847.699951171875
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.5215366705471484e-05,
      "logits/chosen": -2.730821132659912,
      "logits/rejected": -2.7214157581329346,
      "logps/chosen": -302.2289733886719,
      "logps/rejected": -301.8661804199219,
      "loss": 0.3947,
      "pred_label": 1581.550048828125,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -0.503420352935791,
      "rewards/margins": 1.9169738292694092,
      "rewards/rejected": -2.4203941822052,
      "step": 350,
      "use_label": 3940.449951171875
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.463329452852154e-05,
      "logits/chosen": -2.804471969604492,
      "logits/rejected": -2.8228957653045654,
      "logps/chosen": -281.3868103027344,
      "logps/rejected": -285.85137939453125,
      "loss": 0.4073,
      "pred_label": 1646.2750244140625,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -0.8592750430107117,
      "rewards/margins": 1.2057772874832153,
      "rewards/rejected": -2.0650525093078613,
      "step": 360,
      "use_label": 4035.72509765625
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.40512223515716e-05,
      "logits/chosen": -2.7621943950653076,
      "logits/rejected": -2.719569683074951,
      "logps/chosen": -283.62945556640625,
      "logps/rejected": -303.4671936035156,
      "loss": 0.4147,
      "pred_label": 1718.800048828125,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -0.7426743507385254,
      "rewards/margins": 2.031805992126465,
      "rewards/rejected": -2.7744803428649902,
      "step": 370,
      "use_label": 4123.2001953125
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.3469150174621654e-05,
      "logits/chosen": -2.7821342945098877,
      "logits/rejected": -2.7678303718566895,
      "logps/chosen": -292.40106201171875,
      "logps/rejected": -284.78363037109375,
      "loss": 0.3521,
      "pred_label": 1789.875,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.7170799374580383,
      "rewards/margins": 1.70342218875885,
      "rewards/rejected": -2.420502185821533,
      "step": 380,
      "use_label": 4212.125
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.288707799767171e-05,
      "logits/chosen": -2.673171281814575,
      "logits/rejected": -2.665820837020874,
      "logps/chosen": -289.74224853515625,
      "logps/rejected": -273.6763916015625,
      "loss": 0.3695,
      "pred_label": 1873.8499755859375,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.6950327157974243,
      "rewards/margins": 2.244393825531006,
      "rewards/rejected": -2.939426898956299,
      "step": 390,
      "use_label": 4288.14990234375
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.2305005820721776e-05,
      "logits/chosen": -2.809072971343994,
      "logits/rejected": -2.8203823566436768,
      "logps/chosen": -312.44891357421875,
      "logps/rejected": -279.95147705078125,
      "loss": 0.3548,
      "pred_label": 1947.875,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.5397918224334717,
      "rewards/margins": 1.5214416980743408,
      "rewards/rejected": -2.0612335205078125,
      "step": 400,
      "use_label": 4374.125
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.172293364377183e-05,
      "logits/chosen": -2.7990567684173584,
      "logits/rejected": -2.7979464530944824,
      "logps/chosen": -306.15771484375,
      "logps/rejected": -280.4829406738281,
      "loss": 0.3551,
      "pred_label": 2012.9749755859375,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.806006133556366,
      "rewards/margins": 1.3874499797821045,
      "rewards/rejected": -2.1934561729431152,
      "step": 410,
      "use_label": 4469.02490234375
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.1140861466821885e-05,
      "logits/chosen": -2.7010891437530518,
      "logits/rejected": -2.6862704753875732,
      "logps/chosen": -249.2625732421875,
      "logps/rejected": -274.73675537109375,
      "loss": 0.3416,
      "pred_label": 2088.074951171875,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.8682168126106262,
      "rewards/margins": 1.693968415260315,
      "rewards/rejected": -2.562185049057007,
      "step": 420,
      "use_label": 4553.9248046875
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.055878928987195e-05,
      "logits/chosen": -2.734596014022827,
      "logits/rejected": -2.7307052612304688,
      "logps/chosen": -285.8783264160156,
      "logps/rejected": -281.76055908203125,
      "loss": 0.384,
      "pred_label": 2167.85009765625,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.7977985739707947,
      "rewards/margins": 1.426555871963501,
      "rewards/rejected": -2.2243542671203613,
      "step": 430,
      "use_label": 4634.14990234375
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.9976717112922005e-05,
      "logits/chosen": -2.7989730834960938,
      "logits/rejected": -2.8120064735412598,
      "logps/chosen": -291.6087646484375,
      "logps/rejected": -275.2794494628906,
      "loss": 0.3699,
      "pred_label": 2237.574951171875,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.8338401913642883,
      "rewards/margins": 1.3944308757781982,
      "rewards/rejected": -2.228271007537842,
      "step": 440,
      "use_label": 4724.4248046875
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.939464493597206e-05,
      "logits/chosen": -2.5998082160949707,
      "logits/rejected": -2.586780071258545,
      "logps/chosen": -238.42855834960938,
      "logps/rejected": -255.28445434570312,
      "loss": 0.392,
      "pred_label": 2305.5,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.8709820508956909,
      "rewards/margins": 1.692291021347046,
      "rewards/rejected": -2.5632731914520264,
      "step": 450,
      "use_label": 4816.5
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.881257275902212e-05,
      "logits/chosen": -2.800619125366211,
      "logits/rejected": -2.7941062450408936,
      "logps/chosen": -308.8480529785156,
      "logps/rejected": -291.00689697265625,
      "loss": 0.3698,
      "pred_label": 2380.89990234375,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.704586386680603,
      "rewards/margins": 1.4961796998977661,
      "rewards/rejected": -2.200766086578369,
      "step": 460,
      "use_label": 4901.10009765625
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.8230500582072178e-05,
      "logits/chosen": -2.731538772583008,
      "logits/rejected": -2.70729660987854,
      "logps/chosen": -277.3286437988281,
      "logps/rejected": -282.3288269042969,
      "loss": 0.364,
      "pred_label": 2453.449951171875,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.0228168964385986,
      "rewards/margins": 1.525810956954956,
      "rewards/rejected": -2.5486273765563965,
      "step": 470,
      "use_label": 4988.5498046875
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.7648428405122233e-05,
      "logits/chosen": -2.6976053714752197,
      "logits/rejected": -2.6677191257476807,
      "logps/chosen": -255.345703125,
      "logps/rejected": -273.9075012207031,
      "loss": 0.3902,
      "pred_label": 2528.52490234375,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.9771437644958496,
      "rewards/margins": 1.5616614818572998,
      "rewards/rejected": -2.5388054847717285,
      "step": 480,
      "use_label": 5073.47509765625
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.7066356228172297e-05,
      "logits/chosen": -2.7429401874542236,
      "logits/rejected": -2.7300381660461426,
      "logps/chosen": -275.7494201660156,
      "logps/rejected": -261.87689208984375,
      "loss": 0.3725,
      "pred_label": 2596.800048828125,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.8720412254333496,
      "rewards/margins": 1.576229453086853,
      "rewards/rejected": -2.448270559310913,
      "step": 490,
      "use_label": 5165.2001953125
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.6484284051222352e-05,
      "logits/chosen": -2.7378077507019043,
      "logits/rejected": -2.7030460834503174,
      "logps/chosen": -297.50439453125,
      "logps/rejected": -287.47821044921875,
      "loss": 0.3548,
      "pred_label": 2678.425048828125,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -0.8047893643379211,
      "rewards/margins": 2.116807222366333,
      "rewards/rejected": -2.9215962886810303,
      "step": 500,
      "use_label": 5243.5751953125
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.590221187427241e-05,
      "logits/chosen": -2.7614152431488037,
      "logits/rejected": -2.730996608734131,
      "logps/chosen": -285.0855407714844,
      "logps/rejected": -265.8699951171875,
      "loss": 0.3337,
      "pred_label": 2751.699951171875,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -0.9748345613479614,
      "rewards/margins": 2.262641429901123,
      "rewards/rejected": -3.237475872039795,
      "step": 510,
      "use_label": 5330.2998046875
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.532013969732247e-05,
      "logits/chosen": -2.777317523956299,
      "logits/rejected": -2.761885404586792,
      "logps/chosen": -297.0881042480469,
      "logps/rejected": -294.2591247558594,
      "loss": 0.391,
      "pred_label": 2820.324951171875,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.5134023427963257,
      "rewards/margins": 1.941035270690918,
      "rewards/rejected": -2.454437732696533,
      "step": 520,
      "use_label": 5421.6748046875
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.4738067520372525e-05,
      "logits/chosen": -2.779775381088257,
      "logits/rejected": -2.7358291149139404,
      "logps/chosen": -308.1401062011719,
      "logps/rejected": -322.45196533203125,
      "loss": 0.3465,
      "pred_label": 2893.199951171875,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.7824206948280334,
      "rewards/margins": 2.0653204917907715,
      "rewards/rejected": -2.84774112701416,
      "step": 530,
      "use_label": 5508.7998046875
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.4155995343422587e-05,
      "logits/chosen": -2.791782855987549,
      "logits/rejected": -2.7862627506256104,
      "logps/chosen": -296.45611572265625,
      "logps/rejected": -298.04638671875,
      "loss": 0.3581,
      "pred_label": 2980.824951171875,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -0.7919517755508423,
      "rewards/margins": 2.415773868560791,
      "rewards/rejected": -3.2077255249023438,
      "step": 540,
      "use_label": 5581.1748046875
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.3573923166472644e-05,
      "logits/chosen": -2.761887311935425,
      "logits/rejected": -2.735725164413452,
      "logps/chosen": -296.20660400390625,
      "logps/rejected": -290.74456787109375,
      "loss": 0.3915,
      "pred_label": 3058.625,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.7003430128097534,
      "rewards/margins": 2.179884433746338,
      "rewards/rejected": -2.880227565765381,
      "step": 550,
      "use_label": 5663.375
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.2991850989522702e-05,
      "logits/chosen": -2.7279725074768066,
      "logits/rejected": -2.742814540863037,
      "logps/chosen": -283.19342041015625,
      "logps/rejected": -241.63967895507812,
      "loss": 0.3821,
      "pred_label": 3127.85009765625,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.8789056539535522,
      "rewards/margins": 1.5364501476287842,
      "rewards/rejected": -2.415355920791626,
      "step": 560,
      "use_label": 5754.14990234375
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.240977881257276e-05,
      "logits/chosen": -2.757524251937866,
      "logits/rejected": -2.7482261657714844,
      "logps/chosen": -276.85345458984375,
      "logps/rejected": -283.73046875,
      "loss": 0.4022,
      "pred_label": 3200.25,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.736481785774231,
      "rewards/margins": 1.4332786798477173,
      "rewards/rejected": -2.1697604656219482,
      "step": 570,
      "use_label": 5841.75
    },
    {
      "epoch": 0.61,
      "learning_rate": 2.1827706635622818e-05,
      "logits/chosen": -2.7960715293884277,
      "logits/rejected": -2.7829785346984863,
      "logps/chosen": -282.49871826171875,
      "logps/rejected": -286.5024719238281,
      "loss": 0.404,
      "pred_label": 3271.77490234375,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -0.4390278458595276,
      "rewards/margins": 1.7644996643066406,
      "rewards/rejected": -2.2035276889801025,
      "step": 580,
      "use_label": 5930.22509765625
    },
    {
      "epoch": 0.62,
      "learning_rate": 2.124563445867288e-05,
      "logits/chosen": -2.7188072204589844,
      "logits/rejected": -2.70890474319458,
      "logps/chosen": -278.3678894042969,
      "logps/rejected": -304.47906494140625,
      "loss": 0.3951,
      "pred_label": 3332.175048828125,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.6770004034042358,
      "rewards/margins": 1.6928961277008057,
      "rewards/rejected": -2.369896411895752,
      "step": 590,
      "use_label": 6029.8251953125
    },
    {
      "epoch": 0.63,
      "learning_rate": 2.0663562281722934e-05,
      "logits/chosen": -2.788299322128296,
      "logits/rejected": -2.7893383502960205,
      "logps/chosen": -314.5467529296875,
      "logps/rejected": -325.0448913574219,
      "loss": 0.3942,
      "pred_label": 3400.875,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -0.11838535964488983,
      "rewards/margins": 2.110067129135132,
      "rewards/rejected": -2.228452205657959,
      "step": 600,
      "use_label": 6121.125
    },
    {
      "epoch": 0.64,
      "learning_rate": 2.0081490104772992e-05,
      "logits/chosen": -2.7528319358825684,
      "logits/rejected": -2.7486462593078613,
      "logps/chosen": -311.63922119140625,
      "logps/rejected": -263.623046875,
      "loss": 0.3407,
      "pred_label": 3483.35009765625,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.5729966163635254,
      "rewards/margins": 1.8266725540161133,
      "rewards/rejected": -2.3996691703796387,
      "step": 610,
      "use_label": 6198.64990234375
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.9499417927823053e-05,
      "logits/chosen": -2.756082057952881,
      "logits/rejected": -2.706552028656006,
      "logps/chosen": -288.4559631347656,
      "logps/rejected": -281.10784912109375,
      "loss": 0.3528,
      "pred_label": 3560.199951171875,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -0.847445011138916,
      "rewards/margins": 2.0444235801696777,
      "rewards/rejected": -2.891869068145752,
      "step": 620,
      "use_label": 6281.7998046875
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.8917345750873107e-05,
      "logits/chosen": -2.754945993423462,
      "logits/rejected": -2.764570951461792,
      "logps/chosen": -290.9334411621094,
      "logps/rejected": -289.19891357421875,
      "loss": 0.3691,
      "pred_label": 3639.375,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -0.5813703536987305,
      "rewards/margins": 2.221205949783325,
      "rewards/rejected": -2.8025763034820557,
      "step": 630,
      "use_label": 6362.625
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.833527357392317e-05,
      "logits/chosen": -2.7366719245910645,
      "logits/rejected": -2.7163639068603516,
      "logps/chosen": -257.1309509277344,
      "logps/rejected": -274.4014892578125,
      "loss": 0.3542,
      "pred_label": 3707.27490234375,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -0.3939729332923889,
      "rewards/margins": 2.188854217529297,
      "rewards/rejected": -2.582827091217041,
      "step": 640,
      "use_label": 6454.72509765625
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.7753201396973227e-05,
      "logits/chosen": -2.7500858306884766,
      "logits/rejected": -2.72772216796875,
      "logps/chosen": -308.90020751953125,
      "logps/rejected": -307.0010070800781,
      "loss": 0.3495,
      "pred_label": 3787.10009765625,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -0.6951545476913452,
      "rewards/margins": 2.2066218852996826,
      "rewards/rejected": -2.9017765522003174,
      "step": 650,
      "use_label": 6534.89990234375
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.717112922002328e-05,
      "logits/chosen": -2.798872232437134,
      "logits/rejected": -2.7752318382263184,
      "logps/chosen": -304.62677001953125,
      "logps/rejected": -275.2629089355469,
      "loss": 0.3558,
      "pred_label": 3870.0,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.9263059496879578,
      "rewards/margins": 1.612428069114685,
      "rewards/rejected": -2.538734197616577,
      "step": 660,
      "use_label": 6612.0
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.6589057043073342e-05,
      "logits/chosen": -2.7448298931121826,
      "logits/rejected": -2.7179744243621826,
      "logps/chosen": -289.99267578125,
      "logps/rejected": -273.59368896484375,
      "loss": 0.3439,
      "pred_label": 3954.050048828125,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -0.8919633030891418,
      "rewards/margins": 2.3276610374450684,
      "rewards/rejected": -3.2196242809295654,
      "step": 670,
      "use_label": 6687.9501953125
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.60069848661234e-05,
      "logits/chosen": -2.7539925575256348,
      "logits/rejected": -2.757519006729126,
      "logps/chosen": -295.6678771972656,
      "logps/rejected": -280.3077392578125,
      "loss": 0.3793,
      "pred_label": 4029.449951171875,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.8732484579086304,
      "rewards/margins": 1.334531545639038,
      "rewards/rejected": -2.207780122756958,
      "step": 680,
      "use_label": 6772.5498046875
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5424912689173458e-05,
      "logits/chosen": -2.736889123916626,
      "logits/rejected": -2.7280757427215576,
      "logps/chosen": -257.803466796875,
      "logps/rejected": -238.80056762695312,
      "loss": 0.334,
      "pred_label": 4103.75,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -0.7419306039810181,
      "rewards/margins": 2.2096810340881348,
      "rewards/rejected": -2.9516117572784424,
      "step": 690,
      "use_label": 6858.25
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.4842840512223516e-05,
      "logits/chosen": -2.7456727027893066,
      "logits/rejected": -2.7135043144226074,
      "logps/chosen": -286.5834655761719,
      "logps/rejected": -281.8011169433594,
      "loss": 0.323,
      "pred_label": 4186.8251953125,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -1.0893428325653076,
      "rewards/margins": 2.2629120349884033,
      "rewards/rejected": -3.352254867553711,
      "step": 700,
      "use_label": 6935.1748046875
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.4260768335273575e-05,
      "logits/chosen": -2.7708964347839355,
      "logits/rejected": -2.745072603225708,
      "logps/chosen": -298.2213134765625,
      "logps/rejected": -270.7209167480469,
      "loss": 0.3476,
      "pred_label": 4265.2998046875,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.8646717071533203,
      "rewards/margins": 2.080548048019409,
      "rewards/rejected": -2.9452195167541504,
      "step": 710,
      "use_label": 7016.7001953125
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.3678696158323633e-05,
      "logits/chosen": -2.7377257347106934,
      "logits/rejected": -2.7467942237854004,
      "logps/chosen": -283.3219909667969,
      "logps/rejected": -291.44952392578125,
      "loss": 0.2997,
      "pred_label": 4351.2998046875,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -1.1869792938232422,
      "rewards/margins": 2.617690086364746,
      "rewards/rejected": -3.8046698570251465,
      "step": 720,
      "use_label": 7090.7001953125
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.309662398137369e-05,
      "logits/chosen": -2.7232143878936768,
      "logits/rejected": -2.7313239574432373,
      "logps/chosen": -309.1606750488281,
      "logps/rejected": -270.8785095214844,
      "loss": 0.3276,
      "pred_label": 4437.2998046875,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.171083927154541,
      "rewards/margins": 1.749921202659607,
      "rewards/rejected": -2.9210047721862793,
      "step": 730,
      "use_label": 7164.7001953125
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.2514551804423749e-05,
      "logits/chosen": -2.7109317779541016,
      "logits/rejected": -2.722676992416382,
      "logps/chosen": -320.5135192871094,
      "logps/rejected": -325.51531982421875,
      "loss": 0.3404,
      "pred_label": 4520.64990234375,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -1.3141722679138184,
      "rewards/margins": 2.5884618759155273,
      "rewards/rejected": -3.902634382247925,
      "step": 740,
      "use_label": 7241.35009765625
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.1932479627473807e-05,
      "logits/chosen": -2.71235990524292,
      "logits/rejected": -2.6860625743865967,
      "logps/chosen": -276.7895202636719,
      "logps/rejected": -302.22442626953125,
      "loss": 0.3014,
      "pred_label": 4610.0751953125,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -1.6301307678222656,
      "rewards/margins": 2.539576292037964,
      "rewards/rejected": -4.169707298278809,
      "step": 750,
      "use_label": 7311.9248046875
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.1350407450523866e-05,
      "logits/chosen": -2.7227749824523926,
      "logits/rejected": -2.76944899559021,
      "logps/chosen": -306.2732238769531,
      "logps/rejected": -296.8711242675781,
      "loss": 0.323,
      "pred_label": 4696.0751953125,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -1.0810390710830688,
      "rewards/margins": 2.3947594165802,
      "rewards/rejected": -3.4757981300354004,
      "step": 760,
      "use_label": 7385.9248046875
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.0768335273573923e-05,
      "logits/chosen": -2.7785048484802246,
      "logits/rejected": -2.7769179344177246,
      "logps/chosen": -313.1297607421875,
      "logps/rejected": -317.2922058105469,
      "loss": 0.3118,
      "pred_label": 4793.8251953125,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -1.0833934545516968,
      "rewards/margins": 2.1734704971313477,
      "rewards/rejected": -3.256863832473755,
      "step": 770,
      "use_label": 7448.1748046875
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.0186263096623982e-05,
      "logits/chosen": -2.7547028064727783,
      "logits/rejected": -2.7470812797546387,
      "logps/chosen": -294.8021545410156,
      "logps/rejected": -284.2214050292969,
      "loss": 0.3235,
      "pred_label": 4885.39990234375,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.3608357906341553,
      "rewards/margins": 2.3477611541748047,
      "rewards/rejected": -3.708597183227539,
      "step": 780,
      "use_label": 7516.60009765625
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.60419091967404e-06,
      "logits/chosen": -2.7538862228393555,
      "logits/rejected": -2.7440714836120605,
      "logps/chosen": -283.4197998046875,
      "logps/rejected": -299.2707214355469,
      "loss": 0.335,
      "pred_label": 4973.375,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.5610417127609253,
      "rewards/margins": 2.2930331230163574,
      "rewards/rejected": -3.8540749549865723,
      "step": 790,
      "use_label": 7588.625
    },
    {
      "epoch": 0.84,
      "learning_rate": 9.022118742724098e-06,
      "logits/chosen": -2.6376984119415283,
      "logits/rejected": -2.625143051147461,
      "logps/chosen": -329.54010009765625,
      "logps/rejected": -318.21673583984375,
      "loss": 0.3064,
      "pred_label": 5072.7998046875,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -1.5068175792694092,
      "rewards/margins": 2.708643913269043,
      "rewards/rejected": -4.215461254119873,
      "step": 800,
      "use_label": 7649.2001953125
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.440046565774158e-06,
      "logits/chosen": -2.6940813064575195,
      "logits/rejected": -2.6780383586883545,
      "logps/chosen": -297.8315124511719,
      "logps/rejected": -264.34185791015625,
      "loss": 0.3022,
      "pred_label": 5162.1748046875,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.6807565689086914,
      "rewards/margins": 1.872127890586853,
      "rewards/rejected": -3.552884578704834,
      "step": 810,
      "use_label": 7719.8251953125
    },
    {
      "epoch": 0.86,
      "learning_rate": 7.857974388824214e-06,
      "logits/chosen": -2.7031378746032715,
      "logits/rejected": -2.714179277420044,
      "logps/chosen": -305.0399169921875,
      "logps/rejected": -282.8117980957031,
      "loss": 0.3364,
      "pred_label": 5257.125,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -1.1721713542938232,
      "rewards/margins": 2.952846050262451,
      "rewards/rejected": -4.125017166137695,
      "step": 820,
      "use_label": 7784.875
    },
    {
      "epoch": 0.87,
      "learning_rate": 7.275902211874273e-06,
      "logits/chosen": -2.7300140857696533,
      "logits/rejected": -2.728454113006592,
      "logps/chosen": -278.2652282714844,
      "logps/rejected": -319.8196716308594,
      "loss": 0.3174,
      "pred_label": 5347.625,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -1.5451847314834595,
      "rewards/margins": 2.416550636291504,
      "rewards/rejected": -3.961735486984253,
      "step": 830,
      "use_label": 7854.375
    },
    {
      "epoch": 0.88,
      "learning_rate": 6.693830034924331e-06,
      "logits/chosen": -2.716839551925659,
      "logits/rejected": -2.7346701622009277,
      "logps/chosen": -283.65179443359375,
      "logps/rejected": -295.53912353515625,
      "loss": 0.3542,
      "pred_label": 5434.0751953125,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.5634657144546509,
      "rewards/margins": 1.5420477390289307,
      "rewards/rejected": -3.105513334274292,
      "step": 840,
      "use_label": 7927.9248046875
    },
    {
      "epoch": 0.89,
      "learning_rate": 6.111757857974389e-06,
      "logits/chosen": -2.722932815551758,
      "logits/rejected": -2.772580623626709,
      "logps/chosen": -308.41778564453125,
      "logps/rejected": -278.7804260253906,
      "loss": 0.3299,
      "pred_label": 5507.02490234375,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -1.3390918970108032,
      "rewards/margins": 2.2028300762176514,
      "rewards/rejected": -3.541922092437744,
      "step": 850,
      "use_label": 8014.97509765625
    },
    {
      "epoch": 0.9,
      "learning_rate": 5.529685681024447e-06,
      "logits/chosen": -2.7165842056274414,
      "logits/rejected": -2.7188987731933594,
      "logps/chosen": -300.7062683105469,
      "logps/rejected": -310.7410888671875,
      "loss": 0.3504,
      "pred_label": 5592.375,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -1.582289457321167,
      "rewards/margins": 2.2427265644073486,
      "rewards/rejected": -3.8250160217285156,
      "step": 860,
      "use_label": 8089.625
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.947613504074506e-06,
      "logits/chosen": -2.7352089881896973,
      "logits/rejected": -2.74865984916687,
      "logps/chosen": -255.16845703125,
      "logps/rejected": -277.4416198730469,
      "loss": 0.3527,
      "pred_label": 5676.5,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -1.218096137046814,
      "rewards/margins": 2.5011773109436035,
      "rewards/rejected": -3.719273328781128,
      "step": 870,
      "use_label": 8165.5
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.3655413271245635e-06,
      "logits/chosen": -2.754638195037842,
      "logits/rejected": -2.7097582817077637,
      "logps/chosen": -298.7574157714844,
      "logps/rejected": -294.0702209472656,
      "loss": 0.3122,
      "pred_label": 5764.7001953125,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -1.216820240020752,
      "rewards/margins": 2.025149345397949,
      "rewards/rejected": -3.241969585418701,
      "step": 880,
      "use_label": 8237.2998046875
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.7834691501746217e-06,
      "logits/chosen": -2.761935234069824,
      "logits/rejected": -2.767670154571533,
      "logps/chosen": -291.3049011230469,
      "logps/rejected": -307.02984619140625,
      "loss": 0.3179,
      "pred_label": 5856.4248046875,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -1.0020384788513184,
      "rewards/margins": 2.4981894493103027,
      "rewards/rejected": -3.5002281665802,
      "step": 890,
      "use_label": 8305.5751953125
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.2013969732246805e-06,
      "logits/chosen": -2.6637260913848877,
      "logits/rejected": -2.6392276287078857,
      "logps/chosen": -291.3471374511719,
      "logps/rejected": -295.1965637207031,
      "loss": 0.3437,
      "pred_label": 5942.14990234375,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -1.2209066152572632,
      "rewards/margins": 2.44693660736084,
      "rewards/rejected": -3.6678433418273926,
      "step": 900,
      "use_label": 8379.849609375
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.6193247962747383e-06,
      "logits/chosen": -2.7372214794158936,
      "logits/rejected": -2.737588405609131,
      "logps/chosen": -308.132568359375,
      "logps/rejected": -295.11187744140625,
      "loss": 0.3504,
      "pred_label": 6017.4501953125,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -1.17165207862854,
      "rewards/margins": 2.026106834411621,
      "rewards/rejected": -3.1977591514587402,
      "step": 910,
      "use_label": 8464.5498046875
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.037252619324796e-06,
      "logits/chosen": -2.6864240169525146,
      "logits/rejected": -2.693215847015381,
      "logps/chosen": -308.95989990234375,
      "logps/rejected": -296.0351867675781,
      "loss": 0.3214,
      "pred_label": 6108.375,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -1.2072199583053589,
      "rewards/margins": 2.437824249267578,
      "rewards/rejected": -3.6450438499450684,
      "step": 920,
      "use_label": 8533.625
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4551804423748545e-06,
      "logits/chosen": -2.7318947315216064,
      "logits/rejected": -2.6931326389312744,
      "logps/chosen": -274.75,
      "logps/rejected": -280.5450134277344,
      "loss": 0.3077,
      "pred_label": 6194.10009765625,
      "rewards/accuracies": 0.824999988079071,
      "rewards/chosen": -0.6536897420883179,
      "rewards/margins": 3.221295118331909,
      "rewards/rejected": -3.8749847412109375,
      "step": 930,
      "use_label": 8607.900390625
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.731082654249127e-07,
      "logits/chosen": -2.764519214630127,
      "logits/rejected": -2.7424118518829346,
      "logps/chosen": -302.68206787109375,
      "logps/rejected": -300.8379821777344,
      "loss": 0.3294,
      "pred_label": 6278.64990234375,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -1.528961181640625,
      "rewards/margins": 1.8663454055786133,
      "rewards/rejected": -3.3953068256378174,
      "step": 940,
      "use_label": 8683.349609375
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.910360884749709e-07,
      "logits/chosen": -2.7359890937805176,
      "logits/rejected": -2.679305076599121,
      "logps/chosen": -283.5569152832031,
      "logps/rejected": -291.2298889160156,
      "loss": 0.333,
      "pred_label": 6368.97509765625,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -1.5102250576019287,
      "rewards/margins": 2.3179523944854736,
      "rewards/rejected": -3.8281772136688232,
      "step": 950,
      "use_label": 8753.025390625
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -2.742107391357422,
      "eval_logits/rejected": -2.739009141921997,
      "eval_logps/chosen": -295.89080810546875,
      "eval_logps/rejected": -294.5900573730469,
      "eval_loss": 0.3439472019672394,
      "eval_pred_label": 6581.2001953125,
      "eval_rewards/accuracies": 0.7419999837875366,
      "eval_rewards/chosen": -1.163341999053955,
      "eval_rewards/margins": 2.3656928539276123,
      "eval_rewards/rejected": -3.5290346145629883,
      "eval_runtime": 460.9775,
      "eval_samples_per_second": 4.339,
      "eval_steps_per_second": 0.271,
      "eval_use_label": 8950.7998046875,
      "step": 955
    },
    {
      "epoch": 1.0,
      "step": 955,
      "total_flos": 0.0,
      "train_loss": 0.39464539333163756,
      "train_runtime": 25354.6696,
      "train_samples_per_second": 2.411,
      "train_steps_per_second": 0.038
    }
  ],
  "logging_steps": 10,
  "max_steps": 955,
  "num_train_epochs": 1,
  "save_steps": 50,
  "total_flos": 0.0,
  "trial_name": null,
  "trial_params": null
}