{
  "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": 0.10000000149011612,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": 0.02641097828745842,
      "rewards/margins": 0.2079576551914215,
      "rewards/rejected": -0.1815466731786728,
      "step": 40,
      "use_label": 561.9000244140625
    },
    {
      "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": 2.950000047683716,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.06968289613723755,
      "rewards/margins": 0.33043327927589417,
      "rewards/rejected": -0.4001162648200989,
      "step": 50,
      "use_label": 719.0499877929688
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.125e-05,
      "logits/chosen": -2.817157506942749,
      "logits/rejected": -2.8072521686553955,
      "logps/chosen": -300.76080322265625,
      "logps/rejected": -290.3218994140625,
      "loss": 0.5988,
      "pred_label": 11.574999809265137,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.03455673158168793,
      "rewards/margins": 0.3870925307273865,
      "rewards/rejected": -0.4216492772102356,
      "step": 60,
      "use_label": 870.4249877929688
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.6458333333333336e-05,
      "logits/chosen": -2.820021152496338,
      "logits/rejected": -2.813854217529297,
      "logps/chosen": -294.513671875,
      "logps/rejected": -278.9858093261719,
      "loss": 0.5227,
      "pred_label": 25.399999618530273,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.020013216882944107,
      "rewards/margins": 0.5820196866989136,
      "rewards/rejected": -0.6020328998565674,
      "step": 70,
      "use_label": 1016.5999755859375
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.166666666666667e-05,
      "logits/chosen": -2.7645280361175537,
      "logits/rejected": -2.7516016960144043,
      "logps/chosen": -306.9416809082031,
      "logps/rejected": -288.9856262207031,
      "loss": 0.4536,
      "pred_label": 58.599998474121094,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.16363248229026794,
      "rewards/margins": 0.8336677551269531,
      "rewards/rejected": -0.9973002672195435,
      "step": 80,
      "use_label": 1143.4000244140625
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.6875e-05,
      "logits/chosen": -2.6945321559906006,
      "logits/rejected": -2.747893810272217,
      "logps/chosen": -295.63934326171875,
      "logps/rejected": -275.24127197265625,
      "loss": 0.4735,
      "pred_label": 106.5,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": 0.014383295550942421,
      "rewards/margins": 0.911063551902771,
      "rewards/rejected": -0.8966802358627319,
      "step": 90,
      "use_label": 1255.5
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.976717112922003e-05,
      "logits/chosen": -2.719611406326294,
      "logits/rejected": -2.718784809112549,
      "logps/chosen": -276.9577941894531,
      "logps/rejected": -306.4429626464844,
      "loss": 0.484,
      "pred_label": 153.3249969482422,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.28260549902915955,
      "rewards/margins": 1.008049726486206,
      "rewards/rejected": -1.290655255317688,
      "step": 100,
      "use_label": 1368.675048828125
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.918509895227008e-05,
      "logits/chosen": -2.7281861305236816,
      "logits/rejected": -2.687361478805542,
      "logps/chosen": -269.6018981933594,
      "logps/rejected": -263.89166259765625,
      "loss": 0.4361,
      "pred_label": 213.60000610351562,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.4927287697792053,
      "rewards/margins": 0.9790979623794556,
      "rewards/rejected": -1.4718266725540161,
      "step": 110,
      "use_label": 1468.4000244140625
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.860302677532014e-05,
      "logits/chosen": -2.7501158714294434,
      "logits/rejected": -2.742772340774536,
      "logps/chosen": -281.5338439941406,
      "logps/rejected": -259.219970703125,
      "loss": 0.4674,
      "pred_label": 270.0249938964844,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.20209825038909912,
      "rewards/margins": 0.9515292048454285,
      "rewards/rejected": -1.1536273956298828,
      "step": 120,
      "use_label": 1571.9749755859375
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.80209545983702e-05,
      "logits/chosen": -2.8077096939086914,
      "logits/rejected": -2.7647416591644287,
      "logps/chosen": -308.7791748046875,
      "logps/rejected": -282.69976806640625,
      "loss": 0.4803,
      "pred_label": 317.4750061035156,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.15117435157299042,
      "rewards/margins": 0.8076679110527039,
      "rewards/rejected": -0.958842396736145,
      "step": 130,
      "use_label": 1684.5250244140625
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.743888242142026e-05,
      "logits/chosen": -2.814915657043457,
      "logits/rejected": -2.8093409538269043,
      "logps/chosen": -261.60418701171875,
      "logps/rejected": -249.7925567626953,
      "loss": 0.3944,
      "pred_label": 370.67498779296875,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.34973591566085815,
      "rewards/margins": 1.1276283264160156,
      "rewards/rejected": -1.4773643016815186,
      "step": 140,
      "use_label": 1791.324951171875
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.685681024447032e-05,
      "logits/chosen": -2.71696138381958,
      "logits/rejected": -2.740691661834717,
      "logps/chosen": -302.32843017578125,
      "logps/rejected": -298.6430358886719,
      "loss": 0.3948,
      "pred_label": 430.0,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.7442375421524048,
      "rewards/margins": 0.9453207850456238,
      "rewards/rejected": -1.6895582675933838,
      "step": 150,
      "use_label": 1892.0
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.6274738067520374e-05,
      "logits/chosen": -2.7621922492980957,
      "logits/rejected": -2.707430601119995,
      "logps/chosen": -264.4729919433594,
      "logps/rejected": -283.64837646484375,
      "loss": 0.4037,
      "pred_label": 494.9750061035156,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.20909884572029114,
      "rewards/margins": 1.2514889240264893,
      "rewards/rejected": -1.4605878591537476,
      "step": 160,
      "use_label": 1987.0250244140625
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.5692665890570435e-05,
      "logits/chosen": -2.800288438796997,
      "logits/rejected": -2.8005611896514893,
      "logps/chosen": -298.9527282714844,
      "logps/rejected": -264.09014892578125,
      "loss": 0.3923,
      "pred_label": 555.2249755859375,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.24765756726264954,
      "rewards/margins": 1.1968435049057007,
      "rewards/rejected": -1.4445011615753174,
      "step": 170,
      "use_label": 2086.77490234375
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.511059371362049e-05,
      "logits/chosen": -2.7035715579986572,
      "logits/rejected": -2.6763572692871094,
      "logps/chosen": -261.8144836425781,
      "logps/rejected": -294.1943054199219,
      "loss": 0.3547,
      "pred_label": 639.875,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.588528573513031,
      "rewards/margins": 1.7626692056655884,
      "rewards/rejected": -2.3511977195739746,
      "step": 180,
      "use_label": 2162.125
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.452852153667055e-05,
      "logits/chosen": -2.788900375366211,
      "logits/rejected": -2.745344638824463,
      "logps/chosen": -311.311767578125,
      "logps/rejected": -287.326416015625,
      "loss": 0.3357,
      "pred_label": 725.2999877929688,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -0.4055810868740082,
      "rewards/margins": 2.22284197807312,
      "rewards/rejected": -2.6284232139587402,
      "step": 190,
      "use_label": 2236.699951171875
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.394644935972061e-05,
      "logits/chosen": -2.789668560028076,
      "logits/rejected": -2.7655420303344727,
      "logps/chosen": -274.904052734375,
      "logps/rejected": -299.71356201171875,
      "loss": 0.3242,
      "pred_label": 815.0,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.3575278520584106,
      "rewards/margins": 1.7897872924804688,
      "rewards/rejected": -3.147315263748169,
      "step": 200,
      "use_label": 2307.0
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.336437718277067e-05,
      "logits/chosen": -2.7797892093658447,
      "logits/rejected": -2.796938419342041,
      "logps/chosen": -310.7392883300781,
      "logps/rejected": -278.1353454589844,
      "loss": 0.3789,
      "pred_label": 897.2750244140625,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.9491372108459473,
      "rewards/margins": 1.4454370737075806,
      "rewards/rejected": -2.394573926925659,
      "step": 210,
      "use_label": 2384.72509765625
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.278230500582072e-05,
      "logits/chosen": -2.729732036590576,
      "logits/rejected": -2.7413415908813477,
      "logps/chosen": -283.9934387207031,
      "logps/rejected": -297.1932678222656,
      "loss": 0.3046,
      "pred_label": 983.3499755859375,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.9001734852790833,
      "rewards/margins": 1.7534692287445068,
      "rewards/rejected": -2.6536426544189453,
      "step": 220,
      "use_label": 2458.64990234375
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.220023282887078e-05,
      "logits/chosen": -2.7093918323516846,
      "logits/rejected": -2.724216938018799,
      "logps/chosen": -320.80218505859375,
      "logps/rejected": -295.5621643066406,
      "loss": 0.2963,
      "pred_label": 1077.550048828125,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -1.2629055976867676,
      "rewards/margins": 1.8823902606964111,
      "rewards/rejected": -3.1452958583831787,
      "step": 230,
      "use_label": 2524.449951171875
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.161816065192084e-05,
      "logits/chosen": -2.707616090774536,
      "logits/rejected": -2.6788878440856934,
      "logps/chosen": -300.3539123535156,
      "logps/rejected": -273.60333251953125,
      "loss": 0.2668,
      "pred_label": 1176.5250244140625,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -2.5776500701904297,
      "rewards/margins": 2.4807069301605225,
      "rewards/rejected": -5.058356761932373,
      "step": 240,
      "use_label": 2585.47509765625
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.10360884749709e-05,
      "logits/chosen": -2.7368381023406982,
      "logits/rejected": -2.722391366958618,
      "logps/chosen": -316.1106262207031,
      "logps/rejected": -321.246337890625,
      "loss": 0.3166,
      "pred_label": 1270.574951171875,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -2.5470499992370605,
      "rewards/margins": 2.635700225830078,
      "rewards/rejected": -5.182750225067139,
      "step": 250,
      "use_label": 2651.425048828125
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.045401629802096e-05,
      "logits/chosen": -2.7675890922546387,
      "logits/rejected": -2.7713561058044434,
      "logps/chosen": -266.34136962890625,
      "logps/rejected": -286.3191833496094,
      "loss": 0.3408,
      "pred_label": 1354.6500244140625,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -1.7731956243515015,
      "rewards/margins": 1.8298133611679077,
      "rewards/rejected": -3.603008985519409,
      "step": 260,
      "use_label": 2727.35009765625
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.9871944121071014e-05,
      "logits/chosen": -2.764836311340332,
      "logits/rejected": -2.7679736614227295,
      "logps/chosen": -306.4355773925781,
      "logps/rejected": -316.5330505371094,
      "loss": 0.3,
      "pred_label": 1444.324951171875,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -1.4564238786697388,
      "rewards/margins": 2.7216858863830566,
      "rewards/rejected": -4.178110122680664,
      "step": 270,
      "use_label": 2797.675048828125
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.928987194412107e-05,
      "logits/chosen": -2.81805682182312,
      "logits/rejected": -2.8056905269622803,
      "logps/chosen": -309.92047119140625,
      "logps/rejected": -300.94268798828125,
      "loss": 0.3011,
      "pred_label": 1543.4749755859375,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -1.595649003982544,
      "rewards/margins": 2.275697708129883,
      "rewards/rejected": -3.8713467121124268,
      "step": 280,
      "use_label": 2858.52490234375
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.870779976717113e-05,
      "logits/chosen": -2.786515235900879,
      "logits/rejected": -2.7539620399475098,
      "logps/chosen": -312.62060546875,
      "logps/rejected": -268.4746398925781,
      "loss": 0.3045,
      "pred_label": 1641.2249755859375,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -1.9698221683502197,
      "rewards/margins": 1.8559925556182861,
      "rewards/rejected": -3.825814723968506,
      "step": 290,
      "use_label": 2920.77490234375
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.812572759022119e-05,
      "logits/chosen": -2.79560923576355,
      "logits/rejected": -2.744600296020508,
      "logps/chosen": -279.0300598144531,
      "logps/rejected": -275.5624084472656,
      "loss": 0.3083,
      "pred_label": 1731.300048828125,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.5351210832595825,
      "rewards/margins": 2.036700963973999,
      "rewards/rejected": -3.571821928024292,
      "step": 300,
      "use_label": 2990.699951171875
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.7543655413271246e-05,
      "logits/chosen": -2.7574381828308105,
      "logits/rejected": -2.6999497413635254,
      "logps/chosen": -285.3556213378906,
      "logps/rejected": -344.51849365234375,
      "loss": 0.2794,
      "pred_label": 1830.0,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -1.8423988819122314,
      "rewards/margins": 3.7551066875457764,
      "rewards/rejected": -5.597506046295166,
      "step": 310,
      "use_label": 3052.0
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.696158323632131e-05,
      "logits/chosen": -2.7264797687530518,
      "logits/rejected": -2.687101125717163,
      "logps/chosen": -318.2635192871094,
      "logps/rejected": -331.156005859375,
      "loss": 0.2657,
      "pred_label": 1930.050048828125,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -2.228361129760742,
      "rewards/margins": 2.806436777114868,
      "rewards/rejected": -5.034797668457031,
      "step": 320,
      "use_label": 3111.949951171875
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.637951105937136e-05,
      "logits/chosen": -2.7709312438964844,
      "logits/rejected": -2.7176883220672607,
      "logps/chosen": -289.0472717285156,
      "logps/rejected": -277.4143371582031,
      "loss": 0.2868,
      "pred_label": 2034.1500244140625,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.8098747730255127,
      "rewards/margins": 2.50335955619812,
      "rewards/rejected": -4.313233852386475,
      "step": 330,
      "use_label": 3167.85009765625
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.579743888242142e-05,
      "logits/chosen": -2.69682240486145,
      "logits/rejected": -2.7000985145568848,
      "logps/chosen": -308.01141357421875,
      "logps/rejected": -279.4844665527344,
      "loss": 0.3048,
      "pred_label": 2128.949951171875,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.883141279220581,
      "rewards/margins": 1.98026442527771,
      "rewards/rejected": -3.863405704498291,
      "step": 340,
      "use_label": 3233.050048828125
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.5215366705471484e-05,
      "logits/chosen": -2.663508176803589,
      "logits/rejected": -2.6349689960479736,
      "logps/chosen": -314.67962646484375,
      "logps/rejected": -331.1473693847656,
      "loss": 0.2869,
      "pred_label": 2234.77490234375,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": -1.7484877109527588,
      "rewards/margins": 3.6000218391418457,
      "rewards/rejected": -5.348509788513184,
      "step": 350,
      "use_label": 3287.22509765625
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.463329452852154e-05,
      "logits/chosen": -2.7377424240112305,
      "logits/rejected": -2.729678153991699,
      "logps/chosen": -297.26873779296875,
      "logps/rejected": -313.16888427734375,
      "loss": 0.2666,
      "pred_label": 2342.22509765625,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -2.4474658966064453,
      "rewards/margins": 2.349332094192505,
      "rewards/rejected": -4.796797752380371,
      "step": 360,
      "use_label": 3339.77490234375
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.40512223515716e-05,
      "logits/chosen": -2.6696877479553223,
      "logits/rejected": -2.5973799228668213,
      "logps/chosen": -310.88897705078125,
      "logps/rejected": -354.97222900390625,
      "loss": 0.2556,
      "pred_label": 2450.97509765625,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -3.468627452850342,
      "rewards/margins": 4.456360816955566,
      "rewards/rejected": -7.924988746643066,
      "step": 370,
      "use_label": 3391.02490234375
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.3469150174621654e-05,
      "logits/chosen": -2.7021727561950684,
      "logits/rejected": -2.679241418838501,
      "logps/chosen": -320.2778015136719,
      "logps/rejected": -327.36431884765625,
      "loss": 0.2394,
      "pred_label": 2560.375,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -3.5047569274902344,
      "rewards/margins": 3.173809289932251,
      "rewards/rejected": -6.678565979003906,
      "step": 380,
      "use_label": 3441.625
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.288707799767171e-05,
      "logits/chosen": -2.6417362689971924,
      "logits/rejected": -2.6089835166931152,
      "logps/chosen": -295.4016418457031,
      "logps/rejected": -288.3475341796875,
      "loss": 0.2799,
      "pred_label": 2666.675048828125,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -1.2609713077545166,
      "rewards/margins": 3.1455700397491455,
      "rewards/rejected": -4.406540870666504,
      "step": 390,
      "use_label": 3495.324951171875
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.2305005820721776e-05,
      "logits/chosen": -2.774970054626465,
      "logits/rejected": -2.757179021835327,
      "logps/chosen": -317.8477478027344,
      "logps/rejected": -290.2752685546875,
      "loss": 0.2669,
      "pred_label": 2758.64990234375,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.0796750783920288,
      "rewards/margins": 2.0139389038085938,
      "rewards/rejected": -3.093614101409912,
      "step": 400,
      "use_label": 3563.35009765625
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.172293364377183e-05,
      "logits/chosen": -2.777430295944214,
      "logits/rejected": -2.7574374675750732,
      "logps/chosen": -312.574951171875,
      "logps/rejected": -291.488037109375,
      "loss": 0.296,
      "pred_label": 2843.10009765625,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.4477269649505615,
      "rewards/margins": 1.8462398052215576,
      "rewards/rejected": -3.2939670085906982,
      "step": 410,
      "use_label": 3638.89990234375
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.1140861466821885e-05,
      "logits/chosen": -2.718291759490967,
      "logits/rejected": -2.688828706741333,
      "logps/chosen": -254.98593139648438,
      "logps/rejected": -283.18560791015625,
      "loss": 0.3046,
      "pred_label": 2935.14990234375,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.440553069114685,
      "rewards/margins": 1.9665199518203735,
      "rewards/rejected": -3.4070727825164795,
      "step": 420,
      "use_label": 3706.85009765625
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.055878928987195e-05,
      "logits/chosen": -2.7530155181884766,
      "logits/rejected": -2.7446746826171875,
      "logps/chosen": -288.42120361328125,
      "logps/rejected": -287.4293518066406,
      "loss": 0.3446,
      "pred_label": 3031.39990234375,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.0520861148834229,
      "rewards/margins": 1.7391481399536133,
      "rewards/rejected": -2.791234254837036,
      "step": 430,
      "use_label": 3770.60009765625
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.9976717112922005e-05,
      "logits/chosen": -2.8176846504211426,
      "logits/rejected": -2.8231449127197266,
      "logps/chosen": -296.26837158203125,
      "logps/rejected": -284.00335693359375,
      "loss": 0.2945,
      "pred_label": 3121.39990234375,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.2998032569885254,
      "rewards/margins": 1.8008592128753662,
      "rewards/rejected": -3.1006627082824707,
      "step": 440,
      "use_label": 3840.60009765625
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.939464493597206e-05,
      "logits/chosen": -2.6171724796295166,
      "logits/rejected": -2.6079916954040527,
      "logps/chosen": -248.51025390625,
      "logps/rejected": -273.4137268066406,
      "loss": 0.287,
      "pred_label": 3211.925048828125,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -1.8791544437408447,
      "rewards/margins": 2.497044086456299,
      "rewards/rejected": -4.3761982917785645,
      "step": 450,
      "use_label": 3910.074951171875
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.881257275902212e-05,
      "logits/chosen": -2.815080165863037,
      "logits/rejected": -2.8055264949798584,
      "logps/chosen": -318.76666259765625,
      "logps/rejected": -305.4552917480469,
      "loss": 0.3106,
      "pred_label": 3311.449951171875,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.6964454650878906,
      "rewards/margins": 1.949159026145935,
      "rewards/rejected": -3.645604372024536,
      "step": 460,
      "use_label": 3970.550048828125
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.8230500582072178e-05,
      "logits/chosen": -2.736795425415039,
      "logits/rejected": -2.70324444770813,
      "logps/chosen": -289.38824462890625,
      "logps/rejected": -302.501708984375,
      "loss": 0.2652,
      "pred_label": 3412.550048828125,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -2.2287776470184326,
      "rewards/margins": 2.337141513824463,
      "rewards/rejected": -4.565918922424316,
      "step": 470,
      "use_label": 4029.449951171875
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.7648428405122233e-05,
      "logits/chosen": -2.6519317626953125,
      "logits/rejected": -2.616342067718506,
      "logps/chosen": -274.76422119140625,
      "logps/rejected": -303.0647277832031,
      "loss": 0.3052,
      "pred_label": 3516.85009765625,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -2.918997287750244,
      "rewards/margins": 2.5355329513549805,
      "rewards/rejected": -5.454530239105225,
      "step": 480,
      "use_label": 4085.14990234375
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.7066356228172297e-05,
      "logits/chosen": -2.6925861835479736,
      "logits/rejected": -2.67038631439209,
      "logps/chosen": -289.875,
      "logps/rejected": -282.19903564453125,
      "loss": 0.2758,
      "pred_label": 3613.125,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -2.284597873687744,
      "rewards/margins": 2.195889711380005,
      "rewards/rejected": -4.48048734664917,
      "step": 490,
      "use_label": 4148.875
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.6484284051222352e-05,
      "logits/chosen": -2.7040975093841553,
      "logits/rejected": -2.6474618911743164,
      "logps/chosen": -308.2983703613281,
      "logps/rejected": -305.19525146484375,
      "loss": 0.2732,
      "pred_label": 3721.64990234375,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -1.8841865062713623,
      "rewards/margins": 2.809114456176758,
      "rewards/rejected": -4.693300724029541,
      "step": 500,
      "use_label": 4200.35009765625
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.590221187427241e-05,
      "logits/chosen": -2.7423815727233887,
      "logits/rejected": -2.699392795562744,
      "logps/chosen": -296.5220642089844,
      "logps/rejected": -285.8718566894531,
      "loss": 0.2433,
      "pred_label": 3824.625,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -2.1184868812561035,
      "rewards/margins": 3.1191751956939697,
      "rewards/rejected": -5.237661361694336,
      "step": 510,
      "use_label": 4257.375
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.532013969732247e-05,
      "logits/chosen": -2.7525784969329834,
      "logits/rejected": -2.727067708969116,
      "logps/chosen": -304.08343505859375,
      "logps/rejected": -308.287841796875,
      "loss": 0.2989,
      "pred_label": 3925.175048828125,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.2129416465759277,
      "rewards/margins": 2.644366502761841,
      "rewards/rejected": -3.8573079109191895,
      "step": 520,
      "use_label": 4316.8251953125
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.4738067520372525e-05,
      "logits/chosen": -2.759995937347412,
      "logits/rejected": -2.6951136589050293,
      "logps/chosen": -314.9224548339844,
      "logps/rejected": -335.85101318359375,
      "loss": 0.2862,
      "pred_label": 4020.449951171875,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.4606602191925049,
      "rewards/margins": 2.726986885070801,
      "rewards/rejected": -4.187647342681885,
      "step": 530,
      "use_label": 4381.5498046875
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.4155995343422587e-05,
      "logits/chosen": -2.7623682022094727,
      "logits/rejected": -2.743201732635498,
      "logps/chosen": -307.76995849609375,
      "logps/rejected": -319.3795471191406,
      "loss": 0.2676,
      "pred_label": 4129.97509765625,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -1.9233381748199463,
      "rewards/margins": 3.4177041053771973,
      "rewards/rejected": -5.3410420417785645,
      "step": 540,
      "use_label": 4432.02490234375
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.3573923166472644e-05,
      "logits/chosen": -2.766664981842041,
      "logits/rejected": -2.727250576019287,
      "logps/chosen": -307.920166015625,
      "logps/rejected": -313.8031311035156,
      "loss": 0.2619,
      "pred_label": 4236.1748046875,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.871700644493103,
      "rewards/margins": 3.314380645751953,
      "rewards/rejected": -5.186081886291504,
      "step": 550,
      "use_label": 4485.8251953125
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.2991850989522702e-05,
      "logits/chosen": -2.747933864593506,
      "logits/rejected": -2.743129253387451,
      "logps/chosen": -296.66705322265625,
      "logps/rejected": -261.8858947753906,
      "loss": 0.281,
      "pred_label": 4334.72509765625,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -2.2262706756591797,
      "rewards/margins": 2.213707685470581,
      "rewards/rejected": -4.43997859954834,
      "step": 560,
      "use_label": 4547.27490234375
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.240977881257276e-05,
      "logits/chosen": -2.7517714500427246,
      "logits/rejected": -2.7313544750213623,
      "logps/chosen": -288.4844970703125,
      "logps/rejected": -302.5806579589844,
      "loss": 0.2905,
      "pred_label": 4442.875,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.8995840549468994,
      "rewards/margins": 2.1551976203918457,
      "rewards/rejected": -4.054781913757324,
      "step": 570,
      "use_label": 4599.125
    },
    {
      "epoch": 0.61,
      "learning_rate": 2.1827706635622818e-05,
      "logits/chosen": -2.7617430686950684,
      "logits/rejected": -2.73646879196167,
      "logps/chosen": -295.9949645996094,
      "logps/rejected": -310.08331298828125,
      "loss": 0.2821,
      "pred_label": 4553.9501953125,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": -1.7886556386947632,
      "rewards/margins": 2.772958993911743,
      "rewards/rejected": -4.561614513397217,
      "step": 580,
      "use_label": 4648.0498046875
    },
    {
      "epoch": 0.62,
      "learning_rate": 2.124563445867288e-05,
      "logits/chosen": -2.6858975887298584,
      "logits/rejected": -2.6842644214630127,
      "logps/chosen": -292.4759216308594,
      "logps/rejected": -323.04754638671875,
      "loss": 0.2683,
      "pred_label": 4648.1748046875,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -2.0878047943115234,
      "rewards/margins": 2.138942241668701,
      "rewards/rejected": -4.226747035980225,
      "step": 590,
      "use_label": 4713.8251953125
    },
    {
      "epoch": 0.63,
      "learning_rate": 2.0663562281722934e-05,
      "logits/chosen": -2.755965232849121,
      "logits/rejected": -2.7383854389190674,
      "logps/chosen": -327.86962890625,
      "logps/rejected": -351.33074951171875,
      "loss": 0.2585,
      "pred_label": 4749.9501953125,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -1.4506783485412598,
      "rewards/margins": 3.4063594341278076,
      "rewards/rejected": -4.8570380210876465,
      "step": 600,
      "use_label": 4772.0498046875
    },
    {
      "epoch": 0.64,
      "learning_rate": 2.0081490104772992e-05,
      "logits/chosen": -2.7139110565185547,
      "logits/rejected": -2.7096176147460938,
      "logps/chosen": -322.1976013183594,
      "logps/rejected": -281.54266357421875,
      "loss": 0.267,
      "pred_label": 4857.875,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.6288321018218994,
      "rewards/margins": 2.5627994537353516,
      "rewards/rejected": -4.19163179397583,
      "step": 610,
      "use_label": 4824.125
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.9499417927823053e-05,
      "logits/chosen": -2.733909845352173,
      "logits/rejected": -2.6813321113586426,
      "logps/chosen": -296.90032958984375,
      "logps/rejected": -298.9272766113281,
      "loss": 0.2764,
      "pred_label": 4962.77490234375,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": -1.691881775856018,
      "rewards/margins": 2.9819297790527344,
      "rewards/rejected": -4.673811912536621,
      "step": 620,
      "use_label": 4879.22509765625
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.8917345750873107e-05,
      "logits/chosen": -2.7139275074005127,
      "logits/rejected": -2.7132036685943604,
      "logps/chosen": -297.9289855957031,
      "logps/rejected": -306.44940185546875,
      "loss": 0.2715,
      "pred_label": 5066.4501953125,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -1.2809231281280518,
      "rewards/margins": 3.2467029094696045,
      "rewards/rejected": -4.527626991271973,
      "step": 630,
      "use_label": 4935.5498046875
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.833527357392317e-05,
      "logits/chosen": -2.7025070190429688,
      "logits/rejected": -2.669739246368408,
      "logps/chosen": -266.36798095703125,
      "logps/rejected": -293.29669189453125,
      "loss": 0.2907,
      "pred_label": 5163.5,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": -1.3176727294921875,
      "rewards/margins": 3.1546790599823,
      "rewards/rejected": -4.472352027893066,
      "step": 640,
      "use_label": 4998.5
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.7753201396973227e-05,
      "logits/chosen": -2.721750497817993,
      "logits/rejected": -2.6910438537597656,
      "logps/chosen": -316.4061279296875,
      "logps/rejected": -322.0241394042969,
      "loss": 0.2773,
      "pred_label": 5267.2001953125,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -1.4457476139068604,
      "rewards/margins": 2.958343982696533,
      "rewards/rejected": -4.404091835021973,
      "step": 650,
      "use_label": 5054.7998046875
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.717112922002328e-05,
      "logits/chosen": -2.7757656574249268,
      "logits/rejected": -2.7493858337402344,
      "logps/chosen": -311.73321533203125,
      "logps/rejected": -285.9647216796875,
      "loss": 0.2968,
      "pred_label": 5371.77490234375,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.636950135231018,
      "rewards/margins": 1.971967339515686,
      "rewards/rejected": -3.6089179515838623,
      "step": 660,
      "use_label": 5110.22509765625
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.6589057043073342e-05,
      "logits/chosen": -2.7224373817443848,
      "logits/rejected": -2.686645030975342,
      "logps/chosen": -299.3620300292969,
      "logps/rejected": -292.6378173828125,
      "loss": 0.2702,
      "pred_label": 5477.2001953125,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -1.828905701637268,
      "rewards/margins": 3.295133113861084,
      "rewards/rejected": -5.124039173126221,
      "step": 670,
      "use_label": 5164.7998046875
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.60069848661234e-05,
      "logits/chosen": -2.7378199100494385,
      "logits/rejected": -2.7310595512390137,
      "logps/chosen": -304.2184753417969,
      "logps/rejected": -292.1669921875,
      "loss": 0.3204,
      "pred_label": 5576.47509765625,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.7283070087432861,
      "rewards/margins": 1.6653932332992554,
      "rewards/rejected": -3.393700361251831,
      "step": 680,
      "use_label": 5225.52490234375
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5424912689173458e-05,
      "logits/chosen": -2.7175381183624268,
      "logits/rejected": -2.7000772953033447,
      "logps/chosen": -266.01263427734375,
      "logps/rejected": -258.3471374511719,
      "loss": 0.258,
      "pred_label": 5677.625,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -1.56284499168396,
      "rewards/margins": 3.3434224128723145,
      "rewards/rejected": -4.9062676429748535,
      "step": 690,
      "use_label": 5284.375
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.4842840512223516e-05,
      "logits/chosen": -2.735502243041992,
      "logits/rejected": -2.690195322036743,
      "logps/chosen": -293.5663146972656,
      "logps/rejected": -299.68121337890625,
      "loss": 0.2598,
      "pred_label": 5784.4248046875,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -1.7876224517822266,
      "rewards/margins": 3.3526370525360107,
      "rewards/rejected": -5.140259742736816,
      "step": 700,
      "use_label": 5337.5751953125
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.4260768335273575e-05,
      "logits/chosen": -2.7820751667022705,
      "logits/rejected": -2.74092173576355,
      "logps/chosen": -305.2626953125,
      "logps/rejected": -284.5972900390625,
      "loss": 0.2792,
      "pred_label": 5887.0498046875,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -1.5688129663467407,
      "rewards/margins": 2.764045000076294,
      "rewards/rejected": -4.332858085632324,
      "step": 710,
      "use_label": 5394.9501953125
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.3678696158323633e-05,
      "logits/chosen": -2.741055727005005,
      "logits/rejected": -2.739257335662842,
      "logps/chosen": -288.95611572265625,
      "logps/rejected": -306.3753967285156,
      "loss": 0.2461,
      "pred_label": 5995.5498046875,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -1.7503833770751953,
      "rewards/margins": 3.5468757152557373,
      "rewards/rejected": -5.297258377075195,
      "step": 720,
      "use_label": 5446.4501953125
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.309662398137369e-05,
      "logits/chosen": -2.7259597778320312,
      "logits/rejected": -2.719268321990967,
      "logps/chosen": -315.80181884765625,
      "logps/rejected": -282.5977783203125,
      "loss": 0.2265,
      "pred_label": 6100.625,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -1.8351987600326538,
      "rewards/margins": 2.257732629776001,
      "rewards/rejected": -4.092931270599365,
      "step": 730,
      "use_label": 5501.375
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.2514551804423749e-05,
      "logits/chosen": -2.704035758972168,
      "logits/rejected": -2.694396495819092,
      "logps/chosen": -332.0672912597656,
      "logps/rejected": -348.5420837402344,
      "loss": 0.254,
      "pred_label": 6200.5751953125,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -2.469543933868408,
      "rewards/margins": 3.7357654571533203,
      "rewards/rejected": -6.2053093910217285,
      "step": 740,
      "use_label": 5561.4248046875
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.1932479627473807e-05,
      "logits/chosen": -2.717568874359131,
      "logits/rejected": -2.6717875003814697,
      "logps/chosen": -285.0014343261719,
      "logps/rejected": -322.31671142578125,
      "loss": 0.2139,
      "pred_label": 6312.14990234375,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -2.451322078704834,
      "rewards/margins": 3.727611541748047,
      "rewards/rejected": -6.178933620452881,
      "step": 750,
      "use_label": 5609.85009765625
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.1350407450523866e-05,
      "logits/chosen": -2.719574451446533,
      "logits/rejected": -2.7501304149627686,
      "logps/chosen": -310.88677978515625,
      "logps/rejected": -308.3591003417969,
      "loss": 0.2668,
      "pred_label": 6422.97509765625,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": -1.5423939228057861,
      "rewards/margins": 3.0822057723999023,
      "rewards/rejected": -4.624599456787109,
      "step": 760,
      "use_label": 5659.02490234375
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.0768335273573923e-05,
      "logits/chosen": -2.746647357940674,
      "logits/rejected": -2.733283519744873,
      "logps/chosen": -320.37213134765625,
      "logps/rejected": -333.1225280761719,
      "loss": 0.248,
      "pred_label": 6535.22509765625,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -1.807631492614746,
      "rewards/margins": 3.0322647094726562,
      "rewards/rejected": -4.839896202087402,
      "step": 770,
      "use_label": 5706.77490234375
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.0186263096623982e-05,
      "logits/chosen": -2.7445006370544434,
      "logits/rejected": -2.7238264083862305,
      "logps/chosen": -303.70404052734375,
      "logps/rejected": -301.3880615234375,
      "loss": 0.2467,
      "pred_label": 6641.10009765625,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -2.2510251998901367,
      "rewards/margins": 3.1742377281188965,
      "rewards/rejected": -5.425262928009033,
      "step": 780,
      "use_label": 5760.89990234375
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.60419091967404e-06,
      "logits/chosen": -2.7392661571502686,
      "logits/rejected": -2.7158615589141846,
      "logps/chosen": -294.07037353515625,
      "logps/rejected": -322.58758544921875,
      "loss": 0.2523,
      "pred_label": 6745.9501953125,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -2.626100778579712,
      "rewards/margins": 3.5596566200256348,
      "rewards/rejected": -6.185757637023926,
      "step": 790,
      "use_label": 5816.0498046875
    },
    {
      "epoch": 0.84,
      "learning_rate": 9.022118742724098e-06,
      "logits/chosen": -2.6276087760925293,
      "logits/rejected": -2.5911715030670166,
      "logps/chosen": -343.8062438964844,
      "logps/rejected": -343.41070556640625,
      "loss": 0.2253,
      "pred_label": 6867.2998046875,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -2.9334328174591064,
      "rewards/margins": 3.8014297485351562,
      "rewards/rejected": -6.73486328125,
      "step": 800,
      "use_label": 5854.7001953125
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.440046565774158e-06,
      "logits/chosen": -2.6645331382751465,
      "logits/rejected": -2.640127658843994,
      "logps/chosen": -313.86480712890625,
      "logps/rejected": -287.8795471191406,
      "loss": 0.2183,
      "pred_label": 6983.14990234375,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -3.2840819358825684,
      "rewards/margins": 2.622573137283325,
      "rewards/rejected": -5.9066548347473145,
      "step": 810,
      "use_label": 5898.85009765625
    },
    {
      "epoch": 0.86,
      "learning_rate": 7.857974388824214e-06,
      "logits/chosen": -2.684762477874756,
      "logits/rejected": -2.6731903553009033,
      "logps/chosen": -316.68157958984375,
      "logps/rejected": -307.9013671875,
      "loss": 0.2752,
      "pred_label": 7098.10009765625,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": -2.3363418579101562,
      "rewards/margins": 4.297635078430176,
      "rewards/rejected": -6.63397741317749,
      "step": 820,
      "use_label": 5943.89990234375
    },
    {
      "epoch": 0.87,
      "learning_rate": 7.275902211874273e-06,
      "logits/chosen": -2.7039098739624023,
      "logits/rejected": -2.6936678886413574,
      "logps/chosen": -289.11761474609375,
      "logps/rejected": -341.92388916015625,
      "loss": 0.2433,
      "pred_label": 7206.47509765625,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -2.6304259300231934,
      "rewards/margins": 3.5417304039001465,
      "rewards/rejected": -6.172156810760498,
      "step": 830,
      "use_label": 5995.52490234375
    },
    {
      "epoch": 0.88,
      "learning_rate": 6.693830034924331e-06,
      "logits/chosen": -2.6950812339782715,
      "logits/rejected": -2.695385456085205,
      "logps/chosen": -295.2992858886719,
      "logps/rejected": -315.1665954589844,
      "loss": 0.2626,
      "pred_label": 7316.375,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -2.7282135486602783,
      "rewards/margins": 2.3400490283966064,
      "rewards/rejected": -5.068262577056885,
      "step": 840,
      "use_label": 6045.625
    },
    {
      "epoch": 0.89,
      "learning_rate": 6.111757857974389e-06,
      "logits/chosen": -2.691633701324463,
      "logits/rejected": -2.7306087017059326,
      "logps/chosen": -319.0354919433594,
      "logps/rejected": -299.2270812988281,
      "loss": 0.2547,
      "pred_label": 7422.2001953125,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -2.400865316390991,
      "rewards/margins": 3.1857247352600098,
      "rewards/rejected": -5.586589813232422,
      "step": 850,
      "use_label": 6099.7998046875
    },
    {
      "epoch": 0.9,
      "learning_rate": 5.529685681024447e-06,
      "logits/chosen": -2.6894938945770264,
      "logits/rejected": -2.6826186180114746,
      "logps/chosen": -314.03753662109375,
      "logps/rejected": -334.26690673828125,
      "loss": 0.2358,
      "pred_label": 7535.10009765625,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -2.9154160022735596,
      "rewards/margins": 3.2621803283691406,
      "rewards/rejected": -6.177596092224121,
      "step": 860,
      "use_label": 6146.89990234375
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.947613504074506e-06,
      "logits/chosen": -2.6914477348327637,
      "logits/rejected": -2.6861438751220703,
      "logps/chosen": -267.033203125,
      "logps/rejected": -298.70428466796875,
      "loss": 0.2361,
      "pred_label": 7641.60009765625,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -2.404568910598755,
      "rewards/margins": 3.4409728050231934,
      "rewards/rejected": -5.845541477203369,
      "step": 870,
      "use_label": 6200.39990234375
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.3655413271245635e-06,
      "logits/chosen": -2.7052195072174072,
      "logits/rejected": -2.6577229499816895,
      "logps/chosen": -311.93719482421875,
      "logps/rejected": -314.8600769042969,
      "loss": 0.2224,
      "pred_label": 7745.8251953125,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -2.5347952842712402,
      "rewards/margins": 2.786163330078125,
      "rewards/rejected": -5.320958614349365,
      "step": 880,
      "use_label": 6256.1748046875
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.7834691501746217e-06,
      "logits/chosen": -2.718468189239502,
      "logits/rejected": -2.710252523422241,
      "logps/chosen": -304.1473693847656,
      "logps/rejected": -332.4493103027344,
      "loss": 0.2246,
      "pred_label": 7864.375,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -2.2862813472747803,
      "rewards/margins": 3.7558963298797607,
      "rewards/rejected": -6.042177677154541,
      "step": 890,
      "use_label": 6297.625
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.2013969732246805e-06,
      "logits/chosen": -2.6227777004241943,
      "logits/rejected": -2.571324110031128,
      "logps/chosen": -307.7371826171875,
      "logps/rejected": -324.693359375,
      "loss": 0.2493,
      "pred_label": 7978.625,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -2.859910488128662,
      "rewards/margins": 3.757611036300659,
      "rewards/rejected": -6.617520809173584,
      "step": 900,
      "use_label": 6343.375
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.6193247962747383e-06,
      "logits/chosen": -2.701064109802246,
      "logits/rejected": -2.665743350982666,
      "logps/chosen": -323.77337646484375,
      "logps/rejected": -322.24432373046875,
      "loss": 0.238,
      "pred_label": 8087.5,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -2.7357375621795654,
      "rewards/margins": 3.1752612590789795,
      "rewards/rejected": -5.910999298095703,
      "step": 910,
      "use_label": 6394.5
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.037252619324796e-06,
      "logits/chosen": -2.639965772628784,
      "logits/rejected": -2.6299610137939453,
      "logps/chosen": -327.05975341796875,
      "logps/rejected": -325.63909912109375,
      "loss": 0.2258,
      "pred_label": 8202.6748046875,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -3.017204761505127,
      "rewards/margins": 3.5882301330566406,
      "rewards/rejected": -6.605435371398926,
      "step": 920,
      "use_label": 6439.3251953125
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4551804423748545e-06,
      "logits/chosen": -2.69065260887146,
      "logits/rejected": -2.6345713138580322,
      "logps/chosen": -286.16278076171875,
      "logps/rejected": -308.4196472167969,
      "loss": 0.2207,
      "pred_label": 8314.349609375,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": -1.7949655055999756,
      "rewards/margins": 4.867483139038086,
      "rewards/rejected": -6.662448883056641,
      "step": 930,
      "use_label": 6487.64990234375
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.731082654249127e-07,
      "logits/chosen": -2.71934175491333,
      "logits/rejected": -2.67936372756958,
      "logps/chosen": -318.2463073730469,
      "logps/rejected": -326.56634521484375,
      "loss": 0.234,
      "pred_label": 8430.5751953125,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -3.0853824615478516,
      "rewards/margins": 2.8827648162841797,
      "rewards/rejected": -5.968147277832031,
      "step": 940,
      "use_label": 6531.4248046875
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.910360884749709e-07,
      "logits/chosen": -2.6776463985443115,
      "logits/rejected": -2.6152024269104004,
      "logps/chosen": -300.53662109375,
      "logps/rejected": -322.4374694824219,
      "loss": 0.2135,
      "pred_label": 8549.625,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -3.2081990242004395,
      "rewards/margins": 3.740739107131958,
      "rewards/rejected": -6.94893741607666,
      "step": 950,
      "use_label": 6572.375
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -2.6974706649780273,
      "eval_logits/rejected": -2.665019989013672,
      "eval_logps/chosen": -311.6889953613281,
      "eval_logps/rejected": -322.95947265625,
      "eval_loss": 0.2408759742975235,
      "eval_pred_label": 8833.576171875,
      "eval_rewards/accuracies": 0.734000027179718,
      "eval_rewards/chosen": -2.7431609630584717,
      "eval_rewards/margins": 3.6228184700012207,
      "eval_rewards/rejected": -6.36598014831543,
      "eval_runtime": 452.5439,
      "eval_samples_per_second": 4.419,
      "eval_steps_per_second": 0.276,
      "eval_use_label": 6698.423828125,
      "step": 955
    },
    {
      "epoch": 1.0,
      "step": 955,
      "total_flos": 0.0,
      "train_loss": 0.31699458866219243,
      "train_runtime": 25218.7851,
      "train_samples_per_second": 2.424,
      "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
}