{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9993998799759952,
  "eval_steps": 100,
  "global_step": 1249,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 4e-08,
      "logits/chosen": -2.682399272918701,
      "logits/rejected": -2.7047135829925537,
      "logps/chosen": -275.10638427734375,
      "logps/rejected": -271.8466491699219,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.0000000000000003e-07,
      "logits/chosen": -2.606243848800659,
      "logits/rejected": -2.633491277694702,
      "logps/chosen": -301.7389831542969,
      "logps/rejected": -324.2469787597656,
      "loss": 0.6931,
      "rewards/accuracies": 0.4166666567325592,
      "rewards/chosen": 0.00018933300452772528,
      "rewards/margins": 9.413135558133945e-06,
      "rewards/rejected": 0.00017991992353927344,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.000000000000001e-07,
      "logits/chosen": -2.5866377353668213,
      "logits/rejected": -2.5900259017944336,
      "logps/chosen": -269.0643615722656,
      "logps/rejected": -289.1509094238281,
      "loss": 0.6931,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.0011652575340121984,
      "rewards/margins": 6.114605639595538e-05,
      "rewards/rejected": -0.0012264035176485777,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.2000000000000002e-06,
      "logits/chosen": -2.57534122467041,
      "logits/rejected": -2.5880730152130127,
      "logps/chosen": -291.5533752441406,
      "logps/rejected": -311.4080505371094,
      "loss": 0.693,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.0030827566515654325,
      "rewards/margins": 0.0002471129409968853,
      "rewards/rejected": -0.0033298698253929615,
      "step": 30
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6000000000000001e-06,
      "logits/chosen": -2.612988233566284,
      "logits/rejected": -2.619412899017334,
      "logps/chosen": -264.4569396972656,
      "logps/rejected": -273.54107666015625,
      "loss": 0.6927,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.005209661088883877,
      "rewards/margins": 0.0008221397292800248,
      "rewards/rejected": -0.006031800992786884,
      "step": 40
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.0000000000000003e-06,
      "logits/chosen": -2.563633680343628,
      "logits/rejected": -2.5437724590301514,
      "logps/chosen": -264.8982849121094,
      "logps/rejected": -269.68804931640625,
      "loss": 0.6932,
      "rewards/accuracies": 0.44999998807907104,
      "rewards/chosen": -0.008133028633892536,
      "rewards/margins": -1.2002186849713326e-05,
      "rewards/rejected": -0.008121026679873466,
      "step": 50
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.4000000000000003e-06,
      "logits/chosen": -2.6302762031555176,
      "logits/rejected": -2.633596420288086,
      "logps/chosen": -277.2352600097656,
      "logps/rejected": -296.19476318359375,
      "loss": 0.6921,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.013791452161967754,
      "rewards/margins": 0.0020633486565202475,
      "rewards/rejected": -0.01585480198264122,
      "step": 60
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.8000000000000003e-06,
      "logits/chosen": -2.6230902671813965,
      "logits/rejected": -2.6135356426239014,
      "logps/chosen": -280.782958984375,
      "logps/rejected": -286.6590881347656,
      "loss": 0.6931,
      "rewards/accuracies": 0.4937500059604645,
      "rewards/chosen": -0.01885022595524788,
      "rewards/margins": 0.00015832395001780242,
      "rewards/rejected": -0.01900855079293251,
      "step": 70
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.2000000000000003e-06,
      "logits/chosen": -2.6521055698394775,
      "logits/rejected": -2.650635004043579,
      "logps/chosen": -276.92156982421875,
      "logps/rejected": -297.58477783203125,
      "loss": 0.6924,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.017211003229022026,
      "rewards/margins": 0.0015935760457068682,
      "rewards/rejected": -0.018804579973220825,
      "step": 80
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.6000000000000003e-06,
      "logits/chosen": -2.650332450866699,
      "logits/rejected": -2.639291286468506,
      "logps/chosen": -311.17218017578125,
      "logps/rejected": -316.7694091796875,
      "loss": 0.6891,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.017588406801223755,
      "rewards/margins": 0.008481341414153576,
      "rewards/rejected": -0.026069749146699905,
      "step": 90
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.000000000000001e-06,
      "logits/chosen": -2.6498260498046875,
      "logits/rejected": -2.6482200622558594,
      "logps/chosen": -276.51043701171875,
      "logps/rejected": -290.5121154785156,
      "loss": 0.6909,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.02430056408047676,
      "rewards/margins": 0.004733243025839329,
      "rewards/rejected": -0.029033806174993515,
      "step": 100
    },
    {
      "epoch": 0.08,
      "eval_logits/chosen": -2.6246910095214844,
      "eval_logits/rejected": -2.6219358444213867,
      "eval_logps/chosen": -222.40379333496094,
      "eval_logps/rejected": -228.25218200683594,
      "eval_loss": 0.6920604109764099,
      "eval_rewards/accuracies": 0.5426666736602783,
      "eval_rewards/chosen": -0.01691427268087864,
      "eval_rewards/margins": 0.0023173687513917685,
      "eval_rewards/rejected": -0.01923164166510105,
      "eval_runtime": 1621.9667,
      "eval_samples_per_second": 1.846,
      "eval_steps_per_second": 0.231,
      "step": 100
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.4e-06,
      "logits/chosen": -2.6501355171203613,
      "logits/rejected": -2.652731418609619,
      "logps/chosen": -298.9729919433594,
      "logps/rejected": -309.8643493652344,
      "loss": 0.6891,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.025512468069791794,
      "rewards/margins": 0.008652618154883385,
      "rewards/rejected": -0.03416508436203003,
      "step": 110
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.800000000000001e-06,
      "logits/chosen": -2.676758289337158,
      "logits/rejected": -2.6891021728515625,
      "logps/chosen": -278.479736328125,
      "logps/rejected": -296.43212890625,
      "loss": 0.6903,
      "rewards/accuracies": 0.4937500059604645,
      "rewards/chosen": -0.031918395310640335,
      "rewards/margins": 0.006463131867349148,
      "rewards/rejected": -0.03838152438402176,
      "step": 120
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.999755876225375e-06,
      "logits/chosen": -2.645005702972412,
      "logits/rejected": -2.62728214263916,
      "logps/chosen": -294.44366455078125,
      "logps/rejected": -315.0718994140625,
      "loss": 0.6878,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.035983841866254807,
      "rewards/margins": 0.011375428177416325,
      "rewards/rejected": -0.047359269112348557,
      "step": 130
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.997803172081864e-06,
      "logits/chosen": -2.6803853511810303,
      "logits/rejected": -2.680997371673584,
      "logps/chosen": -289.1062927246094,
      "logps/rejected": -302.7191467285156,
      "loss": 0.6855,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.041079964488744736,
      "rewards/margins": 0.016574053093791008,
      "rewards/rejected": -0.057654011994600296,
      "step": 140
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.9938992891651825e-06,
      "logits/chosen": -2.6616640090942383,
      "logits/rejected": -2.6513876914978027,
      "logps/chosen": -277.707763671875,
      "logps/rejected": -300.9044494628906,
      "loss": 0.683,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.05293840169906616,
      "rewards/margins": 0.0216156505048275,
      "rewards/rejected": -0.07455406337976456,
      "step": 150
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.988047277024456e-06,
      "logits/chosen": -2.7210304737091064,
      "logits/rejected": -2.7316393852233887,
      "logps/chosen": -288.5920715332031,
      "logps/rejected": -304.19964599609375,
      "loss": 0.6804,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.07934443652629852,
      "rewards/margins": 0.026776760816574097,
      "rewards/rejected": -0.10612119734287262,
      "step": 160
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.980251707005417e-06,
      "logits/chosen": -2.71783185005188,
      "logits/rejected": -2.690868854522705,
      "logps/chosen": -307.91888427734375,
      "logps/rejected": -315.943359375,
      "loss": 0.689,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.10906956344842911,
      "rewards/margins": 0.009946177713572979,
      "rewards/rejected": -0.11901573836803436,
      "step": 170
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.970518668679459e-06,
      "logits/chosen": -2.729719638824463,
      "logits/rejected": -2.714111804962158,
      "logps/chosen": -304.322998046875,
      "logps/rejected": -311.78167724609375,
      "loss": 0.6816,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.09336166828870773,
      "rewards/margins": 0.0258515365421772,
      "rewards/rejected": -0.11921320110559464,
      "step": 180
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.958855765086722e-06,
      "logits/chosen": -2.757159471511841,
      "logits/rejected": -2.7543435096740723,
      "logps/chosen": -284.23687744140625,
      "logps/rejected": -293.60595703125,
      "loss": 0.6848,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.06431926041841507,
      "rewards/margins": 0.019170444458723068,
      "rewards/rejected": -0.08348970115184784,
      "step": 190
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.945272106796919e-06,
      "logits/chosen": -2.770078420639038,
      "logits/rejected": -2.7745845317840576,
      "logps/chosen": -285.0936584472656,
      "logps/rejected": -300.80975341796875,
      "loss": 0.684,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.0751299113035202,
      "rewards/margins": 0.02216259017586708,
      "rewards/rejected": -0.09729250520467758,
      "step": 200
    },
    {
      "epoch": 0.16,
      "eval_logits/chosen": -2.762911558151245,
      "eval_logits/rejected": -2.759880542755127,
      "eval_logps/chosen": -228.746826171875,
      "eval_logps/rejected": -235.77212524414062,
      "eval_loss": 0.6872997283935547,
      "eval_rewards/accuracies": 0.5566666722297668,
      "eval_rewards/chosen": -0.08034466207027435,
      "eval_rewards/margins": 0.014086335897445679,
      "eval_rewards/rejected": -0.09443099796772003,
      "eval_runtime": 1664.3062,
      "eval_samples_per_second": 1.799,
      "eval_steps_per_second": 0.225,
      "step": 200
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.929778304792537e-06,
      "logits/chosen": -2.7531464099884033,
      "logits/rejected": -2.7568936347961426,
      "logps/chosen": -310.95513916015625,
      "logps/rejected": -315.43353271484375,
      "loss": 0.6749,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.08977816253900528,
      "rewards/margins": 0.044371671974658966,
      "rewards/rejected": -0.13414981961250305,
      "step": 210
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.912386462179987e-06,
      "logits/chosen": -2.7818262577056885,
      "logits/rejected": -2.777902126312256,
      "logps/chosen": -298.80780029296875,
      "logps/rejected": -325.99853515625,
      "loss": 0.6741,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.10158131271600723,
      "rewards/margins": 0.044485487043857574,
      "rewards/rejected": -0.1460667848587036,
      "step": 220
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.893110164735167e-06,
      "logits/chosen": -2.8827590942382812,
      "logits/rejected": -2.8796629905700684,
      "logps/chosen": -305.14068603515625,
      "logps/rejected": -319.39471435546875,
      "loss": 0.6796,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.1311652809381485,
      "rewards/margins": 0.0319681391119957,
      "rewards/rejected": -0.1631334125995636,
      "step": 230
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.871964470290823e-06,
      "logits/chosen": -2.90864634513855,
      "logits/rejected": -2.9260551929473877,
      "logps/chosen": -309.19970703125,
      "logps/rejected": -332.04119873046875,
      "loss": 0.6654,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.1595773994922638,
      "rewards/margins": 0.06538807600736618,
      "rewards/rejected": -0.22496548295021057,
      "step": 240
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.848965896974006e-06,
      "logits/chosen": -2.947906494140625,
      "logits/rejected": -2.940717935562134,
      "logps/chosen": -302.98651123046875,
      "logps/rejected": -325.380859375,
      "loss": 0.6756,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.19427716732025146,
      "rewards/margins": 0.043551910668611526,
      "rewards/rejected": -0.2378290891647339,
      "step": 250
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.8241324103028055e-06,
      "logits/chosen": -3.116504430770874,
      "logits/rejected": -3.088792324066162,
      "logps/chosen": -312.82159423828125,
      "logps/rejected": -328.6378479003906,
      "loss": 0.6614,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.24597156047821045,
      "rewards/margins": 0.07650710642337799,
      "rewards/rejected": -0.322478711605072,
      "step": 260
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.797483409152438e-06,
      "logits/chosen": -3.2219741344451904,
      "logits/rejected": -3.211695432662964,
      "logps/chosen": -308.4544372558594,
      "logps/rejected": -333.4085388183594,
      "loss": 0.6625,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.3319942355155945,
      "rewards/margins": 0.08054188638925552,
      "rewards/rejected": -0.412536084651947,
      "step": 270
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.769039710601669e-06,
      "logits/chosen": -3.368110179901123,
      "logits/rejected": -3.3736987113952637,
      "logps/chosen": -316.3270263671875,
      "logps/rejected": -338.3800048828125,
      "loss": 0.6608,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.42326441407203674,
      "rewards/margins": 0.08332939445972443,
      "rewards/rejected": -0.5065938234329224,
      "step": 280
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.738823533671383e-06,
      "logits/chosen": -3.503385543823242,
      "logits/rejected": -3.490826368331909,
      "logps/chosen": -351.4194030761719,
      "logps/rejected": -368.7189025878906,
      "loss": 0.6789,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.5290869474411011,
      "rewards/margins": 0.048080917447805405,
      "rewards/rejected": -0.5771678686141968,
      "step": 290
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.706858481968017e-06,
      "logits/chosen": -3.464003801345825,
      "logits/rejected": -3.469198226928711,
      "logps/chosen": -340.75830078125,
      "logps/rejected": -352.869384765625,
      "loss": 0.6795,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": -0.6002467274665833,
      "rewards/margins": 0.04778647795319557,
      "rewards/rejected": -0.6480332612991333,
      "step": 300
    },
    {
      "epoch": 0.24,
      "eval_logits/chosen": -3.5198659896850586,
      "eval_logits/rejected": -3.5140511989593506,
      "eval_logps/chosen": -272.0929260253906,
      "eval_logps/rejected": -281.4856262207031,
      "eval_loss": 0.683901846408844,
      "eval_rewards/accuracies": 0.5460000038146973,
      "eval_rewards/chosen": -0.5138051509857178,
      "eval_rewards/margins": 0.037760715931653976,
      "eval_rewards/rejected": -0.551565945148468,
      "eval_runtime": 1688.0776,
      "eval_samples_per_second": 1.774,
      "eval_steps_per_second": 0.222,
      "step": 300
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.673169525245416e-06,
      "logits/chosen": -3.4468257427215576,
      "logits/rejected": -3.422842502593994,
      "logps/chosen": -337.1869812011719,
      "logps/rejected": -369.9530029296875,
      "loss": 0.6606,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.5638504028320312,
      "rewards/margins": 0.09270543605089188,
      "rewards/rejected": -0.6565557718276978,
      "step": 310
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.63778297989952e-06,
      "logits/chosen": -3.598461866378784,
      "logits/rejected": -3.5812110900878906,
      "logps/chosen": -344.33819580078125,
      "logps/rejected": -364.0965270996094,
      "loss": 0.6658,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.5991551280021667,
      "rewards/margins": 0.08249818533658981,
      "rewards/rejected": -0.6816532015800476,
      "step": 320
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.60072648841109e-06,
      "logits/chosen": -3.7547969818115234,
      "logits/rejected": -3.74609637260437,
      "logps/chosen": -364.9189758300781,
      "logps/rejected": -394.4504089355469,
      "loss": 0.6271,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.7271077632904053,
      "rewards/margins": 0.17919641733169556,
      "rewards/rejected": -0.9063041806221008,
      "step": 330
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.562028997752574e-06,
      "logits/chosen": -3.9404075145721436,
      "logits/rejected": -3.9303627014160156,
      "logps/chosen": -380.1604919433594,
      "logps/rejected": -408.8302307128906,
      "loss": 0.676,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.8844378590583801,
      "rewards/margins": 0.08313676714897156,
      "rewards/rejected": -0.9675747156143188,
      "step": 340
    },
    {
      "epoch": 0.28,
      "learning_rate": 4.521720736775947e-06,
      "logits/chosen": -3.974989414215088,
      "logits/rejected": -3.998753786087036,
      "logps/chosen": -397.6668701171875,
      "logps/rejected": -411.4002990722656,
      "loss": 0.6559,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.0197855234146118,
      "rewards/margins": 0.12257959693670273,
      "rewards/rejected": -1.1423652172088623,
      "step": 350
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.479833192599198e-06,
      "logits/chosen": -3.942905902862549,
      "logits/rejected": -3.9193332195281982,
      "logps/chosen": -387.99420166015625,
      "logps/rejected": -410.8706970214844,
      "loss": 0.6613,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.8740938305854797,
      "rewards/margins": 0.10989212989807129,
      "rewards/rejected": -0.9839859008789062,
      "step": 360
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.436399086009928e-06,
      "logits/chosen": -3.781745195388794,
      "logits/rejected": -3.746504306793213,
      "logps/chosen": -363.17657470703125,
      "logps/rejected": -384.13323974609375,
      "loss": 0.6487,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.7560319304466248,
      "rewards/margins": 0.12066509574651718,
      "rewards/rejected": -0.8766969442367554,
      "step": 370
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.391452345905239e-06,
      "logits/chosen": -3.672318696975708,
      "logits/rejected": -3.6834559440612793,
      "logps/chosen": -373.22344970703125,
      "logps/rejected": -389.52777099609375,
      "loss": 0.6611,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.8263392448425293,
      "rewards/margins": 0.11394073814153671,
      "rewards/rejected": -0.9402799606323242,
      "step": 380
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3450280827879125e-06,
      "logits/chosen": -3.7310726642608643,
      "logits/rejected": -3.7564334869384766,
      "logps/chosen": -374.36260986328125,
      "logps/rejected": -394.80010986328125,
      "loss": 0.6607,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.8552727699279785,
      "rewards/margins": 0.11295287311077118,
      "rewards/rejected": -0.9682257771492004,
      "step": 390
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.297162561339554e-06,
      "logits/chosen": -3.6382896900177,
      "logits/rejected": -3.6042380332946777,
      "logps/chosen": -380.48590087890625,
      "logps/rejected": -407.5166015625,
      "loss": 0.6561,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.8749423027038574,
      "rewards/margins": 0.1411461979150772,
      "rewards/rejected": -1.016088604927063,
      "step": 400
    },
    {
      "epoch": 0.32,
      "eval_logits/chosen": -3.7580296993255615,
      "eval_logits/rejected": -3.7484097480773926,
      "eval_logps/chosen": -302.29541015625,
      "eval_logps/rejected": -314.21051025390625,
      "eval_loss": 0.6812021136283875,
      "eval_rewards/accuracies": 0.5573333501815796,
      "eval_rewards/chosen": -0.815830409526825,
      "eval_rewards/margins": 0.06298430263996124,
      "eval_rewards/rejected": -0.8788146376609802,
      "eval_runtime": 1618.6439,
      "eval_samples_per_second": 1.85,
      "eval_steps_per_second": 0.232,
      "step": 400
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.247893172092157e-06,
      "logits/chosen": -3.615405559539795,
      "logits/rejected": -3.612015962600708,
      "logps/chosen": -370.3289794921875,
      "logps/rejected": -405.1226806640625,
      "loss": 0.6581,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.9003459215164185,
      "rewards/margins": 0.13107234239578247,
      "rewards/rejected": -1.0314182043075562,
      "step": 410
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.197258402220187e-06,
      "logits/chosen": -3.65478515625,
      "logits/rejected": -3.6666762828826904,
      "logps/chosen": -379.541748046875,
      "logps/rejected": -418.70880126953125,
      "loss": 0.657,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.9110058546066284,
      "rewards/margins": 0.3442539870738983,
      "rewards/rejected": -1.2552598714828491,
      "step": 420
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.145297805476023e-06,
      "logits/chosen": -3.6817328929901123,
      "logits/rejected": -3.688814640045166,
      "logps/chosen": -376.09783935546875,
      "logps/rejected": -406.4654846191406,
      "loss": 0.6439,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.8991987109184265,
      "rewards/margins": 0.1862173080444336,
      "rewards/rejected": -1.0854160785675049,
      "step": 430
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.092051971292228e-06,
      "logits/chosen": -3.716754198074341,
      "logits/rejected": -3.7106194496154785,
      "logps/chosen": -376.1932067871094,
      "logps/rejected": -401.6561584472656,
      "loss": 0.6554,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.954433798789978,
      "rewards/margins": 0.13505356013774872,
      "rewards/rejected": -1.0894873142242432,
      "step": 440
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.037562493074792e-06,
      "logits/chosen": -3.8129425048828125,
      "logits/rejected": -3.833683729171753,
      "logps/chosen": -403.4329528808594,
      "logps/rejected": -423.4547424316406,
      "loss": 0.6365,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.0463995933532715,
      "rewards/margins": 0.19165393710136414,
      "rewards/rejected": -1.2380534410476685,
      "step": 450
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.981871935712112e-06,
      "logits/chosen": -3.983973741531372,
      "logits/rejected": -3.940070629119873,
      "logps/chosen": -385.3892517089844,
      "logps/rejected": -416.1813049316406,
      "loss": 0.6533,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -1.0447807312011719,
      "rewards/margins": 0.14543746411800385,
      "rewards/rejected": -1.190218210220337,
      "step": 460
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.925023802325094e-06,
      "logits/chosen": -4.042995929718018,
      "logits/rejected": -4.0136399269104,
      "logps/chosen": -401.97381591796875,
      "logps/rejected": -437.3184509277344,
      "loss": 0.6455,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -1.0883817672729492,
      "rewards/margins": 0.2549799978733063,
      "rewards/rejected": -1.343361735343933,
      "step": 470
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.867062500284342e-06,
      "logits/chosen": -4.073556423187256,
      "logits/rejected": -4.043025493621826,
      "logps/chosen": -388.99090576171875,
      "logps/rejected": -425.56787109375,
      "loss": 0.6433,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -1.1277819871902466,
      "rewards/margins": 0.1785212755203247,
      "rewards/rejected": -1.3063032627105713,
      "step": 480
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.8080333065209885e-06,
      "logits/chosen": -4.076624393463135,
      "logits/rejected": -4.084932327270508,
      "logps/chosen": -391.6429138183594,
      "logps/rejected": -392.52362060546875,
      "loss": 0.7064,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -1.1167463064193726,
      "rewards/margins": 0.06357622146606445,
      "rewards/rejected": -1.180322527885437,
      "step": 490
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.7479823321582624e-06,
      "logits/chosen": -3.93993878364563,
      "logits/rejected": -3.9026870727539062,
      "logps/chosen": -378.84918212890625,
      "logps/rejected": -416.9828186035156,
      "loss": 0.633,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.9214572906494141,
      "rewards/margins": 0.19020086526870728,
      "rewards/rejected": -1.1116580963134766,
      "step": 500
    },
    {
      "epoch": 0.4,
      "eval_logits/chosen": -4.107741355895996,
      "eval_logits/rejected": -4.097755432128906,
      "eval_logps/chosen": -310.98577880859375,
      "eval_logps/rejected": -324.42694091796875,
      "eval_loss": 0.678744912147522,
      "eval_rewards/accuracies": 0.5596666932106018,
      "eval_rewards/chosen": -0.9027342796325684,
      "eval_rewards/margins": 0.0782446414232254,
      "eval_rewards/rejected": -0.9809789061546326,
      "eval_runtime": 1619.4632,
      "eval_samples_per_second": 1.849,
      "eval_steps_per_second": 0.232,
      "step": 500
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.686956486491419e-06,
      "logits/chosen": -3.9462807178497314,
      "logits/rejected": -3.941249132156372,
      "logps/chosen": -386.48590087890625,
      "logps/rejected": -424.64508056640625,
      "loss": 0.6313,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.9227803349494934,
      "rewards/margins": 0.29515519738197327,
      "rewards/rejected": -1.217935562133789,
      "step": 510
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.625003440344166e-06,
      "logits/chosen": -4.044493198394775,
      "logits/rejected": -4.073317527770996,
      "logps/chosen": -369.5896301269531,
      "logps/rejected": -382.105224609375,
      "loss": 0.664,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.9153301119804382,
      "rewards/margins": 0.08651997148990631,
      "rewards/rejected": -1.0018501281738281,
      "step": 520
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.562171588830231e-06,
      "logits/chosen": -4.016448497772217,
      "logits/rejected": -3.9958884716033936,
      "logps/chosen": -377.91912841796875,
      "logps/rejected": -404.9534606933594,
      "loss": 0.6943,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.8825477361679077,
      "rewards/margins": 0.0769728347659111,
      "rewards/rejected": -0.9595205187797546,
      "step": 530
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.4985100135491245e-06,
      "logits/chosen": -4.008540630340576,
      "logits/rejected": -3.9679737091064453,
      "logps/chosen": -382.65924072265625,
      "logps/rejected": -425.71954345703125,
      "loss": 0.629,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.8818261027336121,
      "rewards/margins": 0.2214728146791458,
      "rewards/rejected": -1.1032989025115967,
      "step": 540
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.4340684442456673e-06,
      "logits/chosen": -4.049837589263916,
      "logits/rejected": -4.043179035186768,
      "logps/chosen": -384.42938232421875,
      "logps/rejected": -410.5345153808594,
      "loss": 0.6434,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.9648186564445496,
      "rewards/margins": 0.1496874988079071,
      "rewards/rejected": -1.1145063638687134,
      "step": 550
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3688972199631974e-06,
      "logits/chosen": -4.042217254638672,
      "logits/rejected": -4.064842224121094,
      "logps/chosen": -387.6910705566406,
      "logps/rejected": -413.78662109375,
      "loss": 0.6329,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.9935464859008789,
      "rewards/margins": 0.24879872798919678,
      "rewards/rejected": -1.2423454523086548,
      "step": 560
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.3030472497208354e-06,
      "logits/chosen": -4.0646257400512695,
      "logits/rejected": -4.018919944763184,
      "logps/chosen": -385.509765625,
      "logps/rejected": -450.21429443359375,
      "loss": 0.6219,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.0814400911331177,
      "rewards/margins": 0.26284489035606384,
      "rewards/rejected": -1.3442847728729248,
      "step": 570
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.236569972745492e-06,
      "logits/chosen": -4.145129203796387,
      "logits/rejected": -4.1337690353393555,
      "logps/chosen": -380.5148620605469,
      "logps/rejected": -406.1288146972656,
      "loss": 0.6628,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.0649586915969849,
      "rewards/margins": 0.15909543633460999,
      "rewards/rejected": -1.2240540981292725,
      "step": 580
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.1695173182897126e-06,
      "logits/chosen": -4.133418083190918,
      "logits/rejected": -4.11216402053833,
      "logps/chosen": -398.23626708984375,
      "logps/rejected": -435.56646728515625,
      "loss": 0.6554,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.1820625066757202,
      "rewards/margins": 0.16709741950035095,
      "rewards/rejected": -1.349160075187683,
      "step": 590
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.10194166506673e-06,
      "logits/chosen": -4.175902366638184,
      "logits/rejected": -4.1359357833862305,
      "logps/chosen": -390.8631286621094,
      "logps/rejected": -439.28436279296875,
      "loss": 0.6302,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.061226487159729,
      "rewards/margins": 0.2546694874763489,
      "rewards/rejected": -1.3158957958221436,
      "step": 600
    },
    {
      "epoch": 0.48,
      "eval_logits/chosen": -4.443523406982422,
      "eval_logits/rejected": -4.431849002838135,
      "eval_logps/chosen": -337.6354675292969,
      "eval_logps/rejected": -353.24932861328125,
      "eval_loss": 0.6784851551055908,
      "eval_rewards/accuracies": 0.5596666932106018,
      "eval_rewards/chosen": -1.1692306995391846,
      "eval_rewards/margins": 0.09997232258319855,
      "eval_rewards/rejected": -1.2692030668258667,
      "eval_runtime": 1619.3984,
      "eval_samples_per_second": 1.849,
      "eval_steps_per_second": 0.232,
      "step": 600
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.0338958003344115e-06,
      "logits/chosen": -4.3245649337768555,
      "logits/rejected": -4.272718906402588,
      "logps/chosen": -396.521240234375,
      "logps/rejected": -432.1895446777344,
      "loss": 0.6883,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.2242915630340576,
      "rewards/margins": 0.24001073837280273,
      "rewards/rejected": -1.4643023014068604,
      "step": 610
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9654328786600823e-06,
      "logits/chosen": -4.306203365325928,
      "logits/rejected": -4.252989768981934,
      "logps/chosen": -397.3540344238281,
      "logps/rejected": -442.0118103027344,
      "loss": 0.6197,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.2108550071716309,
      "rewards/margins": 0.21409356594085693,
      "rewards/rejected": -1.4249485731124878,
      "step": 620
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.896606380398402e-06,
      "logits/chosen": -4.365767478942871,
      "logits/rejected": -4.406495094299316,
      "logps/chosen": -417.7538146972656,
      "logps/rejected": -443.05548095703125,
      "loss": 0.6623,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.3476839065551758,
      "rewards/margins": 0.18296115100383759,
      "rewards/rejected": -1.5306451320648193,
      "step": 630
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.827470069914772e-06,
      "logits/chosen": -4.2744035720825195,
      "logits/rejected": -4.236593723297119,
      "logps/chosen": -425.37908935546875,
      "logps/rejected": -453.02862548828125,
      "loss": 0.6756,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -1.3940837383270264,
      "rewards/margins": 0.14192768931388855,
      "rewards/rejected": -1.5360115766525269,
      "step": 640
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.7580779535868675e-06,
      "logits/chosen": -4.252664089202881,
      "logits/rejected": -4.254392147064209,
      "logps/chosen": -409.19378662109375,
      "logps/rejected": -438.70556640625,
      "loss": 0.6575,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.230445384979248,
      "rewards/margins": 0.18368306756019592,
      "rewards/rejected": -1.4141284227371216,
      "step": 650
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.688484237617129e-06,
      "logits/chosen": -4.151357650756836,
      "logits/rejected": -4.122767925262451,
      "logps/chosen": -400.3338928222656,
      "logps/rejected": -436.16571044921875,
      "loss": 0.6222,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.108682632446289,
      "rewards/margins": 0.24583642184734344,
      "rewards/rejected": -1.3545191287994385,
      "step": 660
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.6187432856891585e-06,
      "logits/chosen": -4.1051225662231445,
      "logits/rejected": -4.0978288650512695,
      "logps/chosen": -407.8655700683594,
      "logps/rejected": -446.1006774902344,
      "loss": 0.6432,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -1.1575034856796265,
      "rewards/margins": 0.18804897367954254,
      "rewards/rejected": -1.3455523252487183,
      "step": 670
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.548909576501096e-06,
      "logits/chosen": -4.150703430175781,
      "logits/rejected": -4.143389701843262,
      "logps/chosen": -413.3582458496094,
      "logps/rejected": -443.1758728027344,
      "loss": 0.6353,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -1.2182948589324951,
      "rewards/margins": 0.19429844617843628,
      "rewards/rejected": -1.4125932455062866,
      "step": 680
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.4790376612091503e-06,
      "logits/chosen": -4.271695613861084,
      "logits/rejected": -4.229399681091309,
      "logps/chosen": -443.7752990722656,
      "logps/rejected": -475.551025390625,
      "loss": 0.6236,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.456383228302002,
      "rewards/margins": 0.23586714267730713,
      "rewards/rejected": -1.6922504901885986,
      "step": 690
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.40918212081453e-06,
      "logits/chosen": -4.358768939971924,
      "logits/rejected": -4.3066534996032715,
      "logps/chosen": -407.3147888183594,
      "logps/rejected": -466.919921875,
      "loss": 0.5743,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.3695669174194336,
      "rewards/margins": 0.4060749411582947,
      "rewards/rejected": -1.7756417989730835,
      "step": 700
    },
    {
      "epoch": 0.56,
      "eval_logits/chosen": -4.518208026885986,
      "eval_logits/rejected": -4.504719257354736,
      "eval_logps/chosen": -375.0574645996094,
      "eval_logps/rejected": -392.7273254394531,
      "eval_loss": 0.6835331916809082,
      "eval_rewards/accuracies": 0.5630000233650208,
      "eval_rewards/chosen": -1.5434508323669434,
      "eval_rewards/margins": 0.12053229659795761,
      "eval_rewards/rejected": -1.6639831066131592,
      "eval_runtime": 1619.3262,
      "eval_samples_per_second": 1.849,
      "eval_steps_per_second": 0.232,
      "step": 700
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.3393975235270654e-06,
      "logits/chosen": -4.303341865539551,
      "logits/rejected": -4.286491394042969,
      "logps/chosen": -452.05718994140625,
      "logps/rejected": -493.17144775390625,
      "loss": 0.6602,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -1.5600488185882568,
      "rewards/margins": 0.23919770121574402,
      "rewards/rejected": -1.7992465496063232,
      "step": 710
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2697383821388153e-06,
      "logits/chosen": -4.293368816375732,
      "logits/rejected": -4.3109025955200195,
      "logps/chosen": -435.1492614746094,
      "logps/rejected": -460.28033447265625,
      "loss": 0.6504,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -1.4816340208053589,
      "rewards/margins": 0.2186344861984253,
      "rewards/rejected": -1.7002685070037842,
      "step": 720
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2002591114409657e-06,
      "logits/chosen": -4.212637424468994,
      "logits/rejected": -4.208783149719238,
      "logps/chosen": -432.06805419921875,
      "logps/rejected": -468.85601806640625,
      "loss": 0.6325,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.403395652770996,
      "rewards/margins": 0.24927303194999695,
      "rewards/rejected": -1.6526685953140259,
      "step": 730
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.131013985717285e-06,
      "logits/chosen": -4.271391868591309,
      "logits/rejected": -4.220091819763184,
      "logps/chosen": -442.1729431152344,
      "logps/rejected": -491.44384765625,
      "loss": 0.6362,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.4394041299819946,
      "rewards/margins": 0.2614460587501526,
      "rewards/rejected": -1.7008501291275024,
      "step": 740
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.062057096347338e-06,
      "logits/chosen": -4.25800895690918,
      "logits/rejected": -4.223499774932861,
      "logps/chosen": -419.89495849609375,
      "logps/rejected": -435.30389404296875,
      "loss": 0.6593,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.2841438055038452,
      "rewards/margins": 0.15097984671592712,
      "rewards/rejected": -1.4351234436035156,
      "step": 750
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9934423095525733e-06,
      "logits/chosen": -4.121432304382324,
      "logits/rejected": -4.1321306228637695,
      "logps/chosen": -416.6729431152344,
      "logps/rejected": -442.0414123535156,
      "loss": 0.6534,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.1921896934509277,
      "rewards/margins": 0.2546849846839905,
      "rewards/rejected": -1.4468748569488525,
      "step": 760
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9252232243182986e-06,
      "logits/chosen": -4.221813678741455,
      "logits/rejected": -4.169572830200195,
      "logps/chosen": -377.6047668457031,
      "logps/rejected": -425.6900939941406,
      "loss": 0.6026,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.071094036102295,
      "rewards/margins": 0.307754784822464,
      "rewards/rejected": -1.3788487911224365,
      "step": 770
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.8574531305244043e-06,
      "logits/chosen": -4.010577201843262,
      "logits/rejected": -3.9968719482421875,
      "logps/chosen": -418.9974060058594,
      "logps/rejected": -465.41650390625,
      "loss": 0.5956,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.2276278734207153,
      "rewards/margins": 0.3166094124317169,
      "rewards/rejected": -1.5442373752593994,
      "step": 780
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.7901849673175559e-06,
      "logits/chosen": -4.077489376068115,
      "logits/rejected": -4.033568382263184,
      "logps/chosen": -420.52984619140625,
      "logps/rejected": -455.3899841308594,
      "loss": 0.6434,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -1.2396165132522583,
      "rewards/margins": 0.20500020682811737,
      "rewards/rejected": -1.4446165561676025,
      "step": 790
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7234712817573555e-06,
      "logits/chosen": -4.063477516174316,
      "logits/rejected": -4.062304496765137,
      "logps/chosen": -456.8814392089844,
      "logps/rejected": -480.8173828125,
      "loss": 0.6443,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.4422929286956787,
      "rewards/margins": 0.22365431487560272,
      "rewards/rejected": -1.6659473180770874,
      "step": 800
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -4.257167816162109,
      "eval_logits/rejected": -4.245348930358887,
      "eval_logps/chosen": -359.3107604980469,
      "eval_logps/rejected": -377.020751953125,
      "eval_loss": 0.6778839230537415,
      "eval_rewards/accuracies": 0.5666666626930237,
      "eval_rewards/chosen": -1.385983943939209,
      "eval_rewards/margins": 0.12093351036310196,
      "eval_rewards/rejected": -1.5069174766540527,
      "eval_runtime": 1620.4754,
      "eval_samples_per_second": 1.848,
      "eval_steps_per_second": 0.231,
      "step": 800
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.6573641877687936e-06,
      "logits/chosen": -4.076521873474121,
      "logits/rejected": -4.057218074798584,
      "logps/chosen": -422.85699462890625,
      "logps/rejected": -470.95465087890625,
      "loss": 0.6161,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.3746501207351685,
      "rewards/margins": 0.2610599100589752,
      "rewards/rejected": -1.6357100009918213,
      "step": 810
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.591915325433034e-06,
      "logits/chosen": -4.133788108825684,
      "logits/rejected": -4.142486572265625,
      "logps/chosen": -414.0416564941406,
      "logps/rejected": -449.82373046875,
      "loss": 0.6194,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.315294861793518,
      "rewards/margins": 0.30571961402893066,
      "rewards/rejected": -1.6210145950317383,
      "step": 820
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5271758206483664e-06,
      "logits/chosen": -4.143270015716553,
      "logits/rejected": -4.132315158843994,
      "logps/chosen": -438.13037109375,
      "logps/rejected": -471.50030517578125,
      "loss": 0.6481,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.4561512470245361,
      "rewards/margins": 0.22672787308692932,
      "rewards/rejected": -1.682879090309143,
      "step": 830
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4631962451927966e-06,
      "logits/chosen": -4.0487775802612305,
      "logits/rejected": -4.032698631286621,
      "logps/chosen": -431.3470153808594,
      "logps/rejected": -473.2076110839844,
      "loss": 0.6076,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.3846697807312012,
      "rewards/margins": 0.27571359276771545,
      "rewards/rejected": -1.6603834629058838,
      "step": 840
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.4000265772195032e-06,
      "logits/chosen": -4.225982666015625,
      "logits/rejected": -4.171608924865723,
      "logps/chosen": -430.31201171875,
      "logps/rejected": -475.3255920410156,
      "loss": 0.6196,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.3736594915390015,
      "rewards/margins": 0.2828761339187622,
      "rewards/rejected": -1.6565355062484741,
      "step": 850
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3377161622160077e-06,
      "logits/chosen": -4.169137001037598,
      "logits/rejected": -4.160987854003906,
      "logps/chosen": -430.1625061035156,
      "logps/rejected": -470.7685546875,
      "loss": 0.6046,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -1.3943125009536743,
      "rewards/margins": 0.2797546982765198,
      "rewards/rejected": -1.6740672588348389,
      "step": 860
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.276313674457553e-06,
      "logits/chosen": -4.306554794311523,
      "logits/rejected": -4.296151161193848,
      "logps/chosen": -415.35101318359375,
      "logps/rejected": -470.18145751953125,
      "loss": 0.5782,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.413987636566162,
      "rewards/margins": 0.35374554991722107,
      "rewards/rejected": -1.7677332162857056,
      "step": 870
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.2158670789848095e-06,
      "logits/chosen": -4.3886284828186035,
      "logits/rejected": -4.385241508483887,
      "logps/chosen": -460.49005126953125,
      "logps/rejected": -502.52130126953125,
      "loss": 0.608,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.6725581884384155,
      "rewards/margins": 0.34351325035095215,
      "rewards/rejected": -2.016071319580078,
      "step": 880
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.1564235941356016e-06,
      "logits/chosen": -4.544154167175293,
      "logits/rejected": -4.480313301086426,
      "logps/chosen": -452.08245849609375,
      "logps/rejected": -509.763427734375,
      "loss": 0.6428,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -1.7576968669891357,
      "rewards/margins": 0.3635411858558655,
      "rewards/rejected": -2.1212382316589355,
      "step": 890
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.0980296546599254e-06,
      "logits/chosen": -4.359221935272217,
      "logits/rejected": -4.357415199279785,
      "logps/chosen": -476.0204162597656,
      "logps/rejected": -513.69287109375,
      "loss": 0.6651,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.769547462463379,
      "rewards/margins": 0.4242987036705017,
      "rewards/rejected": -2.1938462257385254,
      "step": 900
    },
    {
      "epoch": 0.72,
      "eval_logits/chosen": -4.617808818817139,
      "eval_logits/rejected": -4.603901386260986,
      "eval_logps/chosen": -387.0414123535156,
      "eval_logps/rejected": -406.733154296875,
      "eval_loss": 0.6818779706954956,
      "eval_rewards/accuracies": 0.5693333148956299,
      "eval_rewards/chosen": -1.6632905006408691,
      "eval_rewards/margins": 0.14075076580047607,
      "eval_rewards/rejected": -1.8040413856506348,
      "eval_runtime": 1619.8038,
      "eval_samples_per_second": 1.848,
      "eval_steps_per_second": 0.232,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.040730875447083e-06,
      "logits/chosen": -4.3218793869018555,
      "logits/rejected": -4.3270487785339355,
      "logps/chosen": -451.809814453125,
      "logps/rejected": -478.60491943359375,
      "loss": 0.6503,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.565606713294983,
      "rewards/margins": 0.20372018218040466,
      "rewards/rejected": -1.7693268060684204,
      "step": 910
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.845720158932414e-07,
      "logits/chosen": -4.289103031158447,
      "logits/rejected": -4.304252624511719,
      "logps/chosen": -420.05926513671875,
      "logps/rejected": -442.1366271972656,
      "loss": 0.6505,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.4516870975494385,
      "rewards/margins": 0.20668797194957733,
      "rewards/rejected": -1.6583747863769531,
      "step": 920
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.295969449372796e-07,
      "logits/chosen": -4.317067623138428,
      "logits/rejected": -4.289021968841553,
      "logps/chosen": -417.4593811035156,
      "logps/rejected": -454.826904296875,
      "loss": 0.6219,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.3394687175750732,
      "rewards/margins": 0.25516074895858765,
      "rewards/rejected": -1.5946292877197266,
      "step": 930
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.758486067922176e-07,
      "logits/chosen": -4.306538105010986,
      "logits/rejected": -4.262487411499023,
      "logps/chosen": -414.806884765625,
      "logps/rejected": -465.29931640625,
      "loss": 0.5998,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.3654025793075562,
      "rewards/margins": 0.3171504735946655,
      "rewards/rejected": -1.6825529336929321,
      "step": 940
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.233689873990006e-07,
      "logits/chosen": -4.317531585693359,
      "logits/rejected": -4.282795429229736,
      "logps/chosen": -420.382568359375,
      "logps/rejected": -493.355712890625,
      "loss": 0.5953,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.419424295425415,
      "rewards/margins": 0.5507811903953552,
      "rewards/rejected": -1.970205545425415,
      "step": 950
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.721990816288555e-07,
      "logits/chosen": -4.283775329589844,
      "logits/rejected": -4.2459330558776855,
      "logps/chosen": -395.4678955078125,
      "logps/rejected": -434.7349548339844,
      "loss": 0.6193,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.3240183591842651,
      "rewards/margins": 0.2636045813560486,
      "rewards/rejected": -1.587622880935669,
      "step": 960
    },
    {
      "epoch": 0.78,
      "learning_rate": 7.223788612598148e-07,
      "logits/chosen": -4.310162544250488,
      "logits/rejected": -4.287927627563477,
      "logps/chosen": -436.2978515625,
      "logps/rejected": -470.63079833984375,
      "loss": 0.6316,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.4355452060699463,
      "rewards/margins": 0.18634586036205292,
      "rewards/rejected": -1.6218910217285156,
      "step": 970
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.73947243752448e-07,
      "logits/chosen": -4.377969264984131,
      "logits/rejected": -4.389736652374268,
      "logps/chosen": -458.318359375,
      "logps/rejected": -487.77874755859375,
      "loss": 0.6632,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -1.6293509006500244,
      "rewards/margins": 0.36782675981521606,
      "rewards/rejected": -1.9971777200698853,
      "step": 980
    },
    {
      "epoch": 0.79,
      "learning_rate": 6.269420618491759e-07,
      "logits/chosen": -4.321467876434326,
      "logits/rejected": -4.286978721618652,
      "logps/chosen": -420.07220458984375,
      "logps/rejected": -458.8563537597656,
      "loss": 0.6383,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.4905959367752075,
      "rewards/margins": 0.22982291877269745,
      "rewards/rejected": -1.720418930053711,
      "step": 990
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.814000340209267e-07,
      "logits/chosen": -4.299461364746094,
      "logits/rejected": -4.237879276275635,
      "logps/chosen": -435.0042419433594,
      "logps/rejected": -491.1070861816406,
      "loss": 0.5993,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -1.4882529973983765,
      "rewards/margins": 0.3519567549228668,
      "rewards/rejected": -1.8402099609375,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "eval_logits/chosen": -4.549124717712402,
      "eval_logits/rejected": -4.535641193389893,
      "eval_logps/chosen": -378.4713439941406,
      "eval_logps/rejected": -398.2364196777344,
      "eval_loss": 0.6785325407981873,
      "eval_rewards/accuracies": 0.5683333277702332,
      "eval_rewards/chosen": -1.577589511871338,
      "eval_rewards/margins": 0.1414840668439865,
      "eval_rewards/rejected": -1.7190735340118408,
      "eval_runtime": 1620.6034,
      "eval_samples_per_second": 1.847,
      "eval_steps_per_second": 0.231,
      "step": 1000
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.373567357842111e-07,
      "logits/chosen": -4.278590202331543,
      "logits/rejected": -4.246352195739746,
      "logps/chosen": -429.3436584472656,
      "logps/rejected": -471.84722900390625,
      "loss": 0.5957,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.422374963760376,
      "rewards/margins": 0.33030739426612854,
      "rewards/rejected": -1.7526824474334717,
      "step": 1010
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.948465719110226e-07,
      "logits/chosen": -4.373248100280762,
      "logits/rejected": -4.3430304527282715,
      "logps/chosen": -415.3047790527344,
      "logps/rejected": -445.019287109375,
      "loss": 0.6443,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.40928053855896,
      "rewards/margins": 0.228702574968338,
      "rewards/rejected": -1.6379830837249756,
      "step": 1020
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.539027495532766e-07,
      "logits/chosen": -4.33120059967041,
      "logits/rejected": -4.347836494445801,
      "logps/chosen": -415.25555419921875,
      "logps/rejected": -453.518798828125,
      "loss": 0.6177,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.4035775661468506,
      "rewards/margins": 0.28108319640159607,
      "rewards/rejected": -1.6846606731414795,
      "step": 1030
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.14557252302783e-07,
      "logits/chosen": -4.341530799865723,
      "logits/rejected": -4.305572032928467,
      "logps/chosen": -433.0341796875,
      "logps/rejected": -472.1036682128906,
      "loss": 0.6713,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.5271722078323364,
      "rewards/margins": 0.23339995741844177,
      "rewards/rejected": -1.760572075843811,
      "step": 1040
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.7684081520700884e-07,
      "logits/chosen": -4.2357988357543945,
      "logits/rejected": -4.237625598907471,
      "logps/chosen": -450.54827880859375,
      "logps/rejected": -471.02423095703125,
      "loss": 0.6429,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.5025393962860107,
      "rewards/margins": 0.2623196542263031,
      "rewards/rejected": -1.7648589611053467,
      "step": 1050
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.407829007601507e-07,
      "logits/chosen": -4.270508766174316,
      "logits/rejected": -4.225382328033447,
      "logps/chosen": -428.3861389160156,
      "logps/rejected": -479.0707092285156,
      "loss": 0.6119,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.4203639030456543,
      "rewards/margins": 0.3360464870929718,
      "rewards/rejected": -1.7564103603363037,
      "step": 1060
    },
    {
      "epoch": 0.86,
      "learning_rate": 3.064116758882724e-07,
      "logits/chosen": -4.24053955078125,
      "logits/rejected": -4.1828999519348145,
      "logps/chosen": -443.78485107421875,
      "logps/rejected": -504.28070068359375,
      "loss": 0.6037,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.5132863521575928,
      "rewards/margins": 0.4130435883998871,
      "rewards/rejected": -1.9263302087783813,
      "step": 1070
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.737539899464908e-07,
      "logits/chosen": -4.2971696853637695,
      "logits/rejected": -4.288466453552246,
      "logps/chosen": -403.1819763183594,
      "logps/rejected": -454.8402404785156,
      "loss": 0.6141,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.3817451000213623,
      "rewards/margins": 0.379900187253952,
      "rewards/rejected": -1.7616455554962158,
      "step": 1080
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.4283535374538645e-07,
      "logits/chosen": -4.242150783538818,
      "logits/rejected": -4.240577220916748,
      "logps/chosen": -434.7925720214844,
      "logps/rejected": -473.67803955078125,
      "loss": 0.6122,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.4003090858459473,
      "rewards/margins": 0.31833842396736145,
      "rewards/rejected": -1.7186473608016968,
      "step": 1090
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.1367991962303298e-07,
      "logits/chosen": -4.232297420501709,
      "logits/rejected": -4.2173943519592285,
      "logps/chosen": -418.9490661621094,
      "logps/rejected": -446.4789123535156,
      "loss": 0.6759,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -1.5112884044647217,
      "rewards/margins": 0.1443391889333725,
      "rewards/rejected": -1.6556276082992554,
      "step": 1100
    },
    {
      "epoch": 0.88,
      "eval_logits/chosen": -4.485485553741455,
      "eval_logits/rejected": -4.472198009490967,
      "eval_logps/chosen": -375.86541748046875,
      "eval_logps/rejected": -395.56036376953125,
      "eval_loss": 0.6777821779251099,
      "eval_rewards/accuracies": 0.5686666369438171,
      "eval_rewards/chosen": -1.5515305995941162,
      "eval_rewards/margins": 0.14078289270401,
      "eval_rewards/rejected": -1.6923134326934814,
      "eval_runtime": 1618.2899,
      "eval_samples_per_second": 1.85,
      "eval_steps_per_second": 0.232,
      "step": 1100
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.8631046257820278e-07,
      "logits/chosen": -4.233702659606934,
      "logits/rejected": -4.249814033508301,
      "logps/chosen": -432.20458984375,
      "logps/rejected": -461.7394104003906,
      "loss": 0.6275,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.41855788230896,
      "rewards/margins": 0.2600322365760803,
      "rewards/rejected": -1.678590178489685,
      "step": 1110
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.6074836247950143e-07,
      "logits/chosen": -4.206725597381592,
      "logits/rejected": -4.209759712219238,
      "logps/chosen": -443.52459716796875,
      "logps/rejected": -470.96588134765625,
      "loss": 0.6415,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.4318974018096924,
      "rewards/margins": 0.21563585102558136,
      "rewards/rejected": -1.6475334167480469,
      "step": 1120
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.370135873643097e-07,
      "logits/chosen": -4.226916313171387,
      "logits/rejected": -4.266509532928467,
      "logps/chosen": -436.6336975097656,
      "logps/rejected": -456.3214416503906,
      "loss": 0.6262,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.4590485095977783,
      "rewards/margins": 0.2844700217247009,
      "rewards/rejected": -1.7435184717178345,
      "step": 1130
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.1512467784059372e-07,
      "logits/chosen": -4.302299499511719,
      "logits/rejected": -4.261002540588379,
      "logps/chosen": -398.5301208496094,
      "logps/rejected": -437.83673095703125,
      "loss": 0.6221,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.3698493242263794,
      "rewards/margins": 0.2959776520729065,
      "rewards/rejected": -1.6658270359039307,
      "step": 1140
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.509873260376251e-08,
      "logits/chosen": -4.228875160217285,
      "logits/rejected": -4.17572021484375,
      "logps/chosen": -429.0230407714844,
      "logps/rejected": -501.6388244628906,
      "loss": 0.5814,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.4344505071640015,
      "rewards/margins": 0.4450379014015198,
      "rewards/rejected": -1.8794885873794556,
      "step": 1150
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.695139507988559e-08,
      "logits/chosen": -4.261081218719482,
      "logits/rejected": -4.282492160797119,
      "logps/chosen": -444.05096435546875,
      "logps/rejected": -479.0032653808594,
      "loss": 0.6376,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.457777500152588,
      "rewards/margins": 0.2464013397693634,
      "rewards/rejected": -1.704178810119629,
      "step": 1160
    },
    {
      "epoch": 0.94,
      "learning_rate": 6.069684120570684e-08,
      "logits/chosen": -4.286696434020996,
      "logits/rejected": -4.1916022300720215,
      "logps/chosen": -431.74053955078125,
      "logps/rejected": -482.0506896972656,
      "loss": 0.6355,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.5193744897842407,
      "rewards/margins": 0.3015151619911194,
      "rewards/rejected": -1.8208894729614258,
      "step": 1170
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.634776835499871e-08,
      "logits/chosen": -4.216092109680176,
      "logits/rejected": -4.185781955718994,
      "logps/chosen": -413.5948791503906,
      "logps/rejected": -455.0482482910156,
      "loss": 0.6371,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.4376914501190186,
      "rewards/margins": 0.27367502450942993,
      "rewards/rejected": -1.7113662958145142,
      "step": 1180
    },
    {
      "epoch": 0.95,
      "learning_rate": 3.3915385419908964e-08,
      "logits/chosen": -4.179436206817627,
      "logits/rejected": -4.222240447998047,
      "logps/chosen": -430.4244079589844,
      "logps/rejected": -464.51629638671875,
      "loss": 0.6297,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.4367833137512207,
      "rewards/margins": 0.29136672616004944,
      "rewards/rejected": -1.7281500101089478,
      "step": 1190
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.3409404055043938e-08,
      "logits/chosen": -4.306519508361816,
      "logits/rejected": -4.272242546081543,
      "logps/chosen": -437.791259765625,
      "logps/rejected": -471.58575439453125,
      "loss": 0.6402,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -1.5211617946624756,
      "rewards/margins": 0.2325367033481598,
      "rewards/rejected": -1.7536985874176025,
      "step": 1200
    },
    {
      "epoch": 0.96,
      "eval_logits/chosen": -4.457742691040039,
      "eval_logits/rejected": -4.444427967071533,
      "eval_logps/chosen": -375.10284423828125,
      "eval_logps/rejected": -394.7001647949219,
      "eval_loss": 0.6772644519805908,
      "eval_rewards/accuracies": 0.5690000057220459,
      "eval_rewards/chosen": -1.5439047813415527,
      "eval_rewards/margins": 0.13980673253536224,
      "eval_rewards/rejected": -1.6837116479873657,
      "eval_runtime": 1624.6813,
      "eval_samples_per_second": 1.843,
      "eval_steps_per_second": 0.231,
      "step": 1200
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4838031091134186e-08,
      "logits/chosen": -4.243393898010254,
      "logits/rejected": -4.175347805023193,
      "logps/chosen": -404.5989990234375,
      "logps/rejected": -469.9576110839844,
      "loss": 0.5982,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.3969910144805908,
      "rewards/margins": 0.3869909644126892,
      "rewards/rejected": -1.7839819192886353,
      "step": 1210
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.207962124201774e-09,
      "logits/chosen": -4.263760566711426,
      "logits/rejected": -4.225130081176758,
      "logps/chosen": -435.9114685058594,
      "logps/rejected": -472.66839599609375,
      "loss": 0.6017,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.37319016456604,
      "rewards/margins": 0.3209651708602905,
      "rewards/rejected": -1.6941554546356201,
      "step": 1220
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.5243762852441023e-09,
      "logits/chosen": -4.210858345031738,
      "logits/rejected": -4.163503170013428,
      "logps/chosen": -427.44342041015625,
      "logps/rejected": -469.5209045410156,
      "loss": 0.6376,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.4386152029037476,
      "rewards/margins": 0.2547362446784973,
      "rewards/rejected": -1.6933513879776,
      "step": 1230
    },
    {
      "epoch": 0.99,
      "learning_rate": 7.909321945129278e-10,
      "logits/chosen": -4.18636417388916,
      "logits/rejected": -4.141125679016113,
      "logps/chosen": -443.1625061035156,
      "logps/rejected": -493.43975830078125,
      "loss": 0.6016,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.4015429019927979,
      "rewards/margins": 0.35573890805244446,
      "rewards/rejected": -1.7572818994522095,
      "step": 1240
    },
    {
      "epoch": 1.0,
      "step": 1249,
      "total_flos": 0.0,
      "train_loss": 0.6475976420174225,
      "train_runtime": 42677.4758,
      "train_samples_per_second": 0.469,
      "train_steps_per_second": 0.029
    }
  ],
  "logging_steps": 10,
  "max_steps": 1249,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}