{
  "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": -2.6092891693115234,
      "logits/rejected": -2.629662275314331,
      "logps/chosen": -260.39178466796875,
      "logps/rejected": -256.2431945800781,
      "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": -2.542802095413208,
      "logits/rejected": -2.5701138973236084,
      "logps/chosen": -287.18267822265625,
      "logps/rejected": -308.441162109375,
      "loss": 0.6934,
      "rewards/accuracies": 0.4166666567325592,
      "rewards/chosen": 0.0005982329603284597,
      "rewards/margins": -0.0005221270839683712,
      "rewards/rejected": 0.0011203601025044918,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.000000000000001e-07,
      "logits/chosen": -2.520120859146118,
      "logits/rejected": -2.5234670639038086,
      "logps/chosen": -255.07748413085938,
      "logps/rejected": -273.0997009277344,
      "loss": 0.693,
      "rewards/accuracies": 0.4937500059604645,
      "rewards/chosen": -0.00023935179342515767,
      "rewards/margins": 0.00027643097564578056,
      "rewards/rejected": -0.0005157828563824296,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.2000000000000002e-06,
      "logits/chosen": -2.5063071250915527,
      "logits/rejected": -2.5213770866394043,
      "logps/chosen": -276.6528625488281,
      "logps/rejected": -294.9043884277344,
      "loss": 0.693,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.0008763422374613583,
      "rewards/margins": 0.0003111693658865988,
      "rewards/rejected": -0.001187511719763279,
      "step": 30
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6000000000000001e-06,
      "logits/chosen": -2.542746067047119,
      "logits/rejected": -2.549612522125244,
      "logps/chosen": -251.32003784179688,
      "logps/rejected": -259.71685791015625,
      "loss": 0.6928,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.004003011621534824,
      "rewards/margins": 0.0007247755420394242,
      "rewards/rejected": -0.004727787338197231,
      "step": 40
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.0000000000000003e-06,
      "logits/chosen": -2.4885387420654297,
      "logits/rejected": -2.4705405235290527,
      "logps/chosen": -251.05563354492188,
      "logps/rejected": -255.7452392578125,
      "loss": 0.6933,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.006660168059170246,
      "rewards/margins": -0.00038892580778338015,
      "rewards/rejected": -0.006271242164075375,
      "step": 50
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.4000000000000003e-06,
      "logits/chosen": -2.5540900230407715,
      "logits/rejected": -2.5607895851135254,
      "logps/chosen": -262.1423034667969,
      "logps/rejected": -280.08209228515625,
      "loss": 0.6928,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.009707225486636162,
      "rewards/margins": 0.000684750615619123,
      "rewards/rejected": -0.010391974821686745,
      "step": 60
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.8000000000000003e-06,
      "logits/chosen": -2.5549094676971436,
      "logits/rejected": -2.5427558422088623,
      "logps/chosen": -266.07891845703125,
      "logps/rejected": -271.27740478515625,
      "loss": 0.6926,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.014160098508000374,
      "rewards/margins": 0.0011990683851763606,
      "rewards/rejected": -0.015359166078269482,
      "step": 70
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.2000000000000003e-06,
      "logits/chosen": -2.5750041007995605,
      "logits/rejected": -2.5791172981262207,
      "logps/chosen": -262.36407470703125,
      "logps/rejected": -282.44720458984375,
      "loss": 0.6928,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": -0.015319536440074444,
      "rewards/margins": 0.0008405897533521056,
      "rewards/rejected": -0.01616012677550316,
      "step": 80
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.6000000000000003e-06,
      "logits/chosen": -2.5781426429748535,
      "logits/rejected": -2.5637879371643066,
      "logps/chosen": -295.14898681640625,
      "logps/rejected": -300.16131591796875,
      "loss": 0.6906,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.020102154463529587,
      "rewards/margins": 0.005331115797162056,
      "rewards/rejected": -0.025433272123336792,
      "step": 90
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.000000000000001e-06,
      "logits/chosen": -2.5696635246276855,
      "logits/rejected": -2.569899797439575,
      "logps/chosen": -263.42047119140625,
      "logps/rejected": -276.1015319824219,
      "loss": 0.6917,
      "rewards/accuracies": 0.4937500059604645,
      "rewards/chosen": -0.024685552343726158,
      "rewards/margins": 0.0029262492898851633,
      "rewards/rejected": -0.027611803263425827,
      "step": 100
    },
    {
      "epoch": 0.08,
      "eval_logits/chosen": -2.530292272567749,
      "eval_logits/rejected": -2.5272903442382812,
      "eval_logps/chosen": -210.00074768066406,
      "eval_logps/rejected": -215.47608947753906,
      "eval_loss": 0.6923908591270447,
      "eval_rewards/accuracies": 0.5286666750907898,
      "eval_rewards/chosen": -0.016008317470550537,
      "eval_rewards/margins": 0.001649469486437738,
      "eval_rewards/rejected": -0.01765778660774231,
      "eval_runtime": 1564.6229,
      "eval_samples_per_second": 1.914,
      "eval_steps_per_second": 0.24,
      "step": 100
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.4e-06,
      "logits/chosen": -2.573535680770874,
      "logits/rejected": -2.5764663219451904,
      "logps/chosen": -283.31744384765625,
      "logps/rejected": -293.5372314453125,
      "loss": 0.6918,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": -0.027460157871246338,
      "rewards/margins": 0.0029141721315681934,
      "rewards/rejected": -0.030374327674508095,
      "step": 110
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.800000000000001e-06,
      "logits/chosen": -2.593564033508301,
      "logits/rejected": -2.6072158813476562,
      "logps/chosen": -265.5439758300781,
      "logps/rejected": -282.11444091796875,
      "loss": 0.6911,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": -0.03183265030384064,
      "rewards/margins": 0.004463006276637316,
      "rewards/rejected": -0.036295659840106964,
      "step": 120
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.999755876225375e-06,
      "logits/chosen": -2.559654712677002,
      "logits/rejected": -2.543823480606079,
      "logps/chosen": -279.4471435546875,
      "logps/rejected": -297.8502502441406,
      "loss": 0.6892,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.0379730649292469,
      "rewards/margins": 0.008286955766379833,
      "rewards/rejected": -0.04626002162694931,
      "step": 130
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.997803172081864e-06,
      "logits/chosen": -2.5811429023742676,
      "logits/rejected": -2.581789255142212,
      "logps/chosen": -274.745849609375,
      "logps/rejected": -287.67303466796875,
      "loss": 0.6865,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.04451751708984375,
      "rewards/margins": 0.014026105403900146,
      "rewards/rejected": -0.058543622493743896,
      "step": 140
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.9938992891651825e-06,
      "logits/chosen": -2.545161724090576,
      "logits/rejected": -2.5361251831054688,
      "logps/chosen": -262.94903564453125,
      "logps/rejected": -284.0696716308594,
      "loss": 0.6833,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.0502193346619606,
      "rewards/margins": 0.020780760794878006,
      "rewards/rejected": -0.07100008428096771,
      "step": 150
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.988047277024456e-06,
      "logits/chosen": -2.5855746269226074,
      "logits/rejected": -2.5990371704101562,
      "logps/chosen": -274.5702819824219,
      "logps/rejected": -288.3578796386719,
      "loss": 0.6819,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.07741515338420868,
      "rewards/margins": 0.023703744634985924,
      "rewards/rejected": -0.10111888498067856,
      "step": 160
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.980251707005417e-06,
      "logits/chosen": -2.5723748207092285,
      "logits/rejected": -2.547863006591797,
      "logps/chosen": -291.48736572265625,
      "logps/rejected": -298.7716979980469,
      "loss": 0.6899,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.10616284608840942,
      "rewards/margins": 0.007837113924324512,
      "rewards/rejected": -0.11399996280670166,
      "step": 170
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.970518668679459e-06,
      "logits/chosen": -2.5809338092803955,
      "logits/rejected": -2.566509246826172,
      "logps/chosen": -289.52642822265625,
      "logps/rejected": -295.8729248046875,
      "loss": 0.6849,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.09603477269411087,
      "rewards/margins": 0.01864427700638771,
      "rewards/rejected": -0.11467906087636948,
      "step": 180
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.958855765086722e-06,
      "logits/chosen": -2.608503818511963,
      "logits/rejected": -2.6062302589416504,
      "logps/chosen": -269.84283447265625,
      "logps/rejected": -280.5111083984375,
      "loss": 0.6851,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.05638309568166733,
      "rewards/margins": 0.017940795049071312,
      "rewards/rejected": -0.07432388514280319,
      "step": 190
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.945272106796919e-06,
      "logits/chosen": -2.605029344558716,
      "logits/rejected": -2.6143765449523926,
      "logps/chosen": -270.00830078125,
      "logps/rejected": -285.1636657714844,
      "loss": 0.6854,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.06747416406869888,
      "rewards/margins": 0.018329909071326256,
      "rewards/rejected": -0.08580406755208969,
      "step": 200
    },
    {
      "epoch": 0.16,
      "eval_logits/chosen": -2.57635235786438,
      "eval_logits/rejected": -2.573589563369751,
      "eval_logps/chosen": -215.42254638671875,
      "eval_logps/rejected": -222.06101989746094,
      "eval_loss": 0.6875421404838562,
      "eval_rewards/accuracies": 0.5563333630561829,
      "eval_rewards/chosen": -0.07022587954998016,
      "eval_rewards/margins": 0.013281372375786304,
      "eval_rewards/rejected": -0.08350725471973419,
      "eval_runtime": 1564.649,
      "eval_samples_per_second": 1.914,
      "eval_steps_per_second": 0.24,
      "step": 200
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.929778304792537e-06,
      "logits/chosen": -2.594109058380127,
      "logits/rejected": -2.5997750759124756,
      "logps/chosen": -294.9931640625,
      "logps/rejected": -299.47515869140625,
      "loss": 0.6783,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.08973054587841034,
      "rewards/margins": 0.03516639396548271,
      "rewards/rejected": -0.12489694356918335,
      "step": 210
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.912386462179987e-06,
      "logits/chosen": -2.6223044395446777,
      "logits/rejected": -2.6194934844970703,
      "logps/chosen": -282.9168701171875,
      "logps/rejected": -308.5869140625,
      "loss": 0.6757,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.08990589529275894,
      "rewards/margins": 0.039798181504011154,
      "rewards/rejected": -0.1297040730714798,
      "step": 220
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.893110164735167e-06,
      "logits/chosen": -2.686192512512207,
      "logits/rejected": -2.6861042976379395,
      "logps/chosen": -288.02984619140625,
      "logps/rejected": -301.3773193359375,
      "loss": 0.6819,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.11503495275974274,
      "rewards/margins": 0.02641122415661812,
      "rewards/rejected": -0.14144618809223175,
      "step": 230
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.871964470290823e-06,
      "logits/chosen": -2.660379409790039,
      "logits/rejected": -2.6843981742858887,
      "logps/chosen": -289.6697998046875,
      "logps/rejected": -310.08245849609375,
      "loss": 0.6712,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.12107042968273163,
      "rewards/margins": 0.05008827522397041,
      "rewards/rejected": -0.17115873098373413,
      "step": 240
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.848965896974006e-06,
      "logits/chosen": -2.6900529861450195,
      "logits/rejected": -2.691603183746338,
      "logps/chosen": -281.68792724609375,
      "logps/rejected": -301.64129638671875,
      "loss": 0.6784,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.1229969710111618,
      "rewards/margins": 0.03473496437072754,
      "rewards/rejected": -0.15773193538188934,
      "step": 250
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.8241324103028055e-06,
      "logits/chosen": -2.794003963470459,
      "logits/rejected": -2.756747007369995,
      "logps/chosen": -288.9064025878906,
      "logps/rejected": -300.8970947265625,
      "loss": 0.6737,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.1542636603116989,
      "rewards/margins": 0.044557444751262665,
      "rewards/rejected": -0.19882109761238098,
      "step": 260
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.797483409152438e-06,
      "logits/chosen": -2.8427796363830566,
      "logits/rejected": -2.8422060012817383,
      "logps/chosen": -280.02099609375,
      "logps/rejected": -300.798583984375,
      "loss": 0.667,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.17782548069953918,
      "rewards/margins": 0.0627431869506836,
      "rewards/rejected": -0.24056866765022278,
      "step": 270
    },
    {
      "epoch": 0.22,
      "learning_rate": 4.769039710601669e-06,
      "logits/chosen": -2.926100015640259,
      "logits/rejected": -2.9281368255615234,
      "logps/chosen": -280.8794250488281,
      "logps/rejected": -298.9492492675781,
      "loss": 0.6732,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.2086031138896942,
      "rewards/margins": 0.047397904098033905,
      "rewards/rejected": -0.25600099563598633,
      "step": 280
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.738823533671383e-06,
      "logits/chosen": -2.972533702850342,
      "logits/rejected": -2.9721083641052246,
      "logps/chosen": -306.55865478515625,
      "logps/rejected": -322.1519470214844,
      "loss": 0.6769,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.24026934802532196,
      "rewards/margins": 0.041927777230739594,
      "rewards/rejected": -0.28219714760780334,
      "step": 290
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.706858481968017e-06,
      "logits/chosen": -2.922761917114258,
      "logits/rejected": -2.9264841079711914,
      "logps/chosen": -295.0879821777344,
      "logps/rejected": -304.8949279785156,
      "loss": 0.682,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.2886089086532593,
      "rewards/margins": 0.03215286135673523,
      "rewards/rejected": -0.3207617700099945,
      "step": 300
    },
    {
      "epoch": 0.24,
      "eval_logits/chosen": -2.920912981033325,
      "eval_logits/rejected": -2.9179892539978027,
      "eval_logps/chosen": -232.2801055908203,
      "eval_logps/rejected": -240.21971130371094,
      "eval_loss": 0.684112012386322,
      "eval_rewards/accuracies": 0.5450000166893005,
      "eval_rewards/chosen": -0.23880183696746826,
      "eval_rewards/margins": 0.026292217895388603,
      "eval_rewards/rejected": -0.2650940418243408,
      "eval_runtime": 1563.5865,
      "eval_samples_per_second": 1.915,
      "eval_steps_per_second": 0.24,
      "step": 300
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.673169525245416e-06,
      "logits/chosen": -2.912137269973755,
      "logits/rejected": -2.890232563018799,
      "logps/chosen": -291.8182678222656,
      "logps/rejected": -320.7854309082031,
      "loss": 0.6652,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.257565438747406,
      "rewards/margins": 0.06874646991491318,
      "rewards/rejected": -0.3263119161128998,
      "step": 310
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.63778297989952e-06,
      "logits/chosen": -3.007861375808716,
      "logits/rejected": -2.9914684295654297,
      "logps/chosen": -297.81695556640625,
      "logps/rejected": -314.2324523925781,
      "loss": 0.6678,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.28555119037628174,
      "rewards/margins": 0.06271664053201675,
      "rewards/rejected": -0.3482677936553955,
      "step": 320
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.60072648841109e-06,
      "logits/chosen": -3.0622572898864746,
      "logits/rejected": -3.0570547580718994,
      "logps/chosen": -311.02105712890625,
      "logps/rejected": -333.46630859375,
      "loss": 0.6444,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.3358006477355957,
      "rewards/margins": 0.11871840804815292,
      "rewards/rejected": -0.454519122838974,
      "step": 330
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.562028997752574e-06,
      "logits/chosen": -3.153225898742676,
      "logits/rejected": -3.1641342639923096,
      "logps/chosen": -311.9708557128906,
      "logps/rejected": -338.3099670410156,
      "loss": 0.6682,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.3542831838130951,
      "rewards/margins": 0.068769671022892,
      "rewards/rejected": -0.4230528473854065,
      "step": 340
    },
    {
      "epoch": 0.28,
      "learning_rate": 4.521720736775947e-06,
      "logits/chosen": -3.2916831970214844,
      "logits/rejected": -3.3004555702209473,
      "logps/chosen": -328.3070373535156,
      "logps/rejected": -338.16192626953125,
      "loss": 0.6564,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.4777124524116516,
      "rewards/margins": 0.09214212745428085,
      "rewards/rejected": -0.5698545575141907,
      "step": 350
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.479833192599198e-06,
      "logits/chosen": -3.4253318309783936,
      "logits/rejected": -3.4066455364227295,
      "logps/chosen": -329.5577697753906,
      "logps/rejected": -350.79742431640625,
      "loss": 0.6555,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.46095147728919983,
      "rewards/margins": 0.09699266403913498,
      "rewards/rejected": -0.5579441785812378,
      "step": 360
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.436399086009928e-06,
      "logits/chosen": -3.4043822288513184,
      "logits/rejected": -3.3838889598846436,
      "logps/chosen": -317.8493957519531,
      "logps/rejected": -334.6202087402344,
      "loss": 0.6576,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.4469628930091858,
      "rewards/margins": 0.08755813539028168,
      "rewards/rejected": -0.5345209836959839,
      "step": 370
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.391452345905239e-06,
      "logits/chosen": -3.347275972366333,
      "logits/rejected": -3.35723614692688,
      "logps/chosen": -325.1240234375,
      "logps/rejected": -339.3655090332031,
      "loss": 0.6618,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.4934774935245514,
      "rewards/margins": 0.09260674566030502,
      "rewards/rejected": -0.586084246635437,
      "step": 380
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3450280827879125e-06,
      "logits/chosen": -3.4295692443847656,
      "logits/rejected": -3.4535210132598877,
      "logps/chosen": -323.19415283203125,
      "logps/rejected": -340.07989501953125,
      "loss": 0.6639,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.4877980351448059,
      "rewards/margins": 0.08489425480365753,
      "rewards/rejected": -0.5726922154426575,
      "step": 390
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.297162561339554e-06,
      "logits/chosen": -3.366345167160034,
      "logits/rejected": -3.3470451831817627,
      "logps/chosen": -330.50311279296875,
      "logps/rejected": -352.4704895019531,
      "loss": 0.6634,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.5246509313583374,
      "rewards/margins": 0.09757552295923233,
      "rewards/rejected": -0.6222264766693115,
      "step": 400
    },
    {
      "epoch": 0.32,
      "eval_logits/chosen": -3.460310935974121,
      "eval_logits/rejected": -3.4548885822296143,
      "eval_logps/chosen": -256.72369384765625,
      "eval_logps/rejected": -266.585693359375,
      "eval_loss": 0.681212306022644,
      "eval_rewards/accuracies": 0.5486666560173035,
      "eval_rewards/chosen": -0.4832378029823303,
      "eval_rewards/margins": 0.04551612213253975,
      "eval_rewards/rejected": -0.528753936290741,
      "eval_runtime": 1563.8204,
      "eval_samples_per_second": 1.915,
      "eval_steps_per_second": 0.24,
      "step": 400
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.247893172092157e-06,
      "logits/chosen": -3.3930275440216064,
      "logits/rejected": -3.3995296955108643,
      "logps/chosen": -320.64154052734375,
      "logps/rejected": -351.81414794921875,
      "loss": 0.6576,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.5477547645568848,
      "rewards/margins": 0.10663916915655136,
      "rewards/rejected": -0.6543939709663391,
      "step": 410
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.197258402220187e-06,
      "logits/chosen": -3.4664406776428223,
      "logits/rejected": -3.4772841930389404,
      "logps/chosen": -328.4302673339844,
      "logps/rejected": -347.1799011230469,
      "loss": 0.6516,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.5463472604751587,
      "rewards/margins": 0.14043554663658142,
      "rewards/rejected": -0.686782717704773,
      "step": 420
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.145297805476023e-06,
      "logits/chosen": -3.5236573219299316,
      "logits/rejected": -3.5294857025146484,
      "logps/chosen": -326.1194763183594,
      "logps/rejected": -350.4049987792969,
      "loss": 0.6477,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.5520918965339661,
      "rewards/margins": 0.13327525556087494,
      "rewards/rejected": -0.685367226600647,
      "step": 430
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.092051971292228e-06,
      "logits/chosen": -3.597280502319336,
      "logits/rejected": -3.597381591796875,
      "logps/chosen": -324.05584716796875,
      "logps/rejected": -346.5114440917969,
      "loss": 0.654,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.5706614255905151,
      "rewards/margins": 0.11499401181936264,
      "rewards/rejected": -0.6856554746627808,
      "step": 440
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.037562493074792e-06,
      "logits/chosen": -3.742100238800049,
      "logits/rejected": -3.7566616535186768,
      "logps/chosen": -349.8996887207031,
      "logps/rejected": -364.50506591796875,
      "loss": 0.6477,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.6562848091125488,
      "rewards/margins": 0.1411934792995453,
      "rewards/rejected": -0.7974783182144165,
      "step": 450
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.981871935712112e-06,
      "logits/chosen": -3.968726396560669,
      "logits/rejected": -3.941209077835083,
      "logps/chosen": -332.73443603515625,
      "logps/rejected": -360.47625732421875,
      "loss": 0.6493,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -0.6563684344291687,
      "rewards/margins": 0.13294601440429688,
      "rewards/rejected": -0.7893144488334656,
      "step": 460
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.925023802325094e-06,
      "logits/chosen": -4.149798393249512,
      "logits/rejected": -4.129707336425781,
      "logps/chosen": -354.22802734375,
      "logps/rejected": -380.85040283203125,
      "loss": 0.6505,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.7849351763725281,
      "rewards/margins": 0.16557727754116058,
      "rewards/rejected": -0.9505125284194946,
      "step": 470
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.867062500284342e-06,
      "logits/chosen": -4.189931392669678,
      "logits/rejected": -4.1646270751953125,
      "logps/chosen": -344.6732482910156,
      "logps/rejected": -376.7868957519531,
      "loss": 0.647,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.8410800099372864,
      "rewards/margins": 0.15282076597213745,
      "rewards/rejected": -0.9939007759094238,
      "step": 480
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.8080333065209885e-06,
      "logits/chosen": -4.1885223388671875,
      "logits/rejected": -4.1956305503845215,
      "logps/chosen": -351.62237548828125,
      "logps/rejected": -351.5168151855469,
      "loss": 0.6998,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.8627797961235046,
      "rewards/margins": 0.05857674032449722,
      "rewards/rejected": -0.9213564991950989,
      "step": 490
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.7479823321582624e-06,
      "logits/chosen": -4.04768705368042,
      "logits/rejected": -4.022109031677246,
      "logps/chosen": -345.1391296386719,
      "logps/rejected": -380.88482666015625,
      "loss": 0.6296,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.7372187972068787,
      "rewards/margins": 0.17785954475402832,
      "rewards/rejected": -0.915078341960907,
      "step": 500
    },
    {
      "epoch": 0.4,
      "eval_logits/chosen": -4.174871921539307,
      "eval_logits/rejected": -4.1668291091918945,
      "eval_logps/chosen": -277.3629150390625,
      "eval_logps/rejected": -289.35430908203125,
      "eval_loss": 0.678155779838562,
      "eval_rewards/accuracies": 0.5600000023841858,
      "eval_rewards/chosen": -0.6896299123764038,
      "eval_rewards/margins": 0.06681026518344879,
      "eval_rewards/rejected": -0.7564401626586914,
      "eval_runtime": 1565.1126,
      "eval_samples_per_second": 1.913,
      "eval_steps_per_second": 0.24,
      "step": 500
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.686956486491419e-06,
      "logits/chosen": -4.035499572753906,
      "logits/rejected": -4.026083469390869,
      "logps/chosen": -354.2256774902344,
      "logps/rejected": -383.2754821777344,
      "loss": 0.642,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.7423269748687744,
      "rewards/margins": 0.21402092278003693,
      "rewards/rejected": -0.9563478231430054,
      "step": 510
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.625003440344166e-06,
      "logits/chosen": -4.110401630401611,
      "logits/rejected": -4.130579948425293,
      "logps/chosen": -333.2506103515625,
      "logps/rejected": -345.42529296875,
      "loss": 0.6662,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.6982457637786865,
      "rewards/margins": 0.08196356892585754,
      "rewards/rejected": -0.7802093625068665,
      "step": 520
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.562171588830231e-06,
      "logits/chosen": -4.092467308044434,
      "logits/rejected": -4.074677467346191,
      "logps/chosen": -342.26513671875,
      "logps/rejected": -370.6748352050781,
      "loss": 0.6627,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.6801224946975708,
      "rewards/margins": 0.11174317449331284,
      "rewards/rejected": -0.7918656468391418,
      "step": 530
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.4985100135491245e-06,
      "logits/chosen": -4.104280471801758,
      "logits/rejected": -4.07126522064209,
      "logps/chosen": -353.3872375488281,
      "logps/rejected": -391.28289794921875,
      "loss": 0.6383,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.7574983239173889,
      "rewards/margins": 0.18485240638256073,
      "rewards/rejected": -0.9423506855964661,
      "step": 540
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.4340684442456673e-06,
      "logits/chosen": -4.108157634735107,
      "logits/rejected": -4.100740432739258,
      "logps/chosen": -353.9938049316406,
      "logps/rejected": -375.70562744140625,
      "loss": 0.6498,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.8029891848564148,
      "rewards/margins": 0.12438172101974487,
      "rewards/rejected": -0.9273708462715149,
      "step": 550
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3688972199631974e-06,
      "logits/chosen": -4.0825605392456055,
      "logits/rejected": -4.094653129577637,
      "logps/chosen": -352.5472412109375,
      "logps/rejected": -372.22125244140625,
      "loss": 0.646,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.7980318069458008,
      "rewards/margins": 0.1836821585893631,
      "rewards/rejected": -0.9817139506340027,
      "step": 560
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.3030472497208354e-06,
      "logits/chosen": -4.089273452758789,
      "logits/rejected": -4.051713943481445,
      "logps/chosen": -349.65447998046875,
      "logps/rejected": -411.3336486816406,
      "loss": 0.6145,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.8661645650863647,
      "rewards/margins": 0.2521691918373108,
      "rewards/rejected": -1.1183335781097412,
      "step": 570
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.236569972745492e-06,
      "logits/chosen": -4.18710994720459,
      "logits/rejected": -4.17570161819458,
      "logps/chosen": -347.78521728515625,
      "logps/rejected": -368.5164489746094,
      "loss": 0.6671,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.878218948841095,
      "rewards/margins": 0.11807211488485336,
      "rewards/rejected": -0.9962911605834961,
      "step": 580
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.1695173182897126e-06,
      "logits/chosen": -4.180907249450684,
      "logits/rejected": -4.169510841369629,
      "logps/chosen": -363.77081298828125,
      "logps/rejected": -398.118408203125,
      "loss": 0.6542,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.982130229473114,
      "rewards/margins": 0.1635296642780304,
      "rewards/rejected": -1.1456598043441772,
      "step": 590
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.10194166506673e-06,
      "logits/chosen": -4.227340221405029,
      "logits/rejected": -4.191034317016602,
      "logps/chosen": -362.04327392578125,
      "logps/rejected": -401.2042541503906,
      "loss": 0.6503,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.9192866086959839,
      "rewards/margins": 0.18541966378688812,
      "rewards/rejected": -1.1047064065933228,
      "step": 600
    },
    {
      "epoch": 0.48,
      "eval_logits/chosen": -4.443284511566162,
      "eval_logits/rejected": -4.434456825256348,
      "eval_logps/chosen": -304.2834167480469,
      "eval_logps/rejected": -318.1134033203125,
      "eval_loss": 0.6769664287567139,
      "eval_rewards/accuracies": 0.5533333420753479,
      "eval_rewards/chosen": -0.9588348269462585,
      "eval_rewards/margins": 0.0851958841085434,
      "eval_rewards/rejected": -1.0440306663513184,
      "eval_runtime": 1565.9018,
      "eval_samples_per_second": 1.912,
      "eval_steps_per_second": 0.239,
      "step": 600
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.0338958003344115e-06,
      "logits/chosen": -4.3410210609436035,
      "logits/rejected": -4.294773101806641,
      "logps/chosen": -360.5263977050781,
      "logps/rejected": -391.7284240722656,
      "loss": 0.6595,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.9942121505737305,
      "rewards/margins": 0.21516022086143494,
      "rewards/rejected": -1.2093722820281982,
      "step": 610
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9654328786600823e-06,
      "logits/chosen": -4.326909065246582,
      "logits/rejected": -4.2857208251953125,
      "logps/chosen": -359.14569091796875,
      "logps/rejected": -399.8367004394531,
      "loss": 0.627,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.9758028984069824,
      "rewards/margins": 0.18271294236183167,
      "rewards/rejected": -1.1585158109664917,
      "step": 620
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.896606380398402e-06,
      "logits/chosen": -4.396684169769287,
      "logits/rejected": -4.430912971496582,
      "logps/chosen": -377.33551025390625,
      "logps/rejected": -398.4137878417969,
      "loss": 0.6688,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -1.0871975421905518,
      "rewards/margins": 0.13647660613059998,
      "rewards/rejected": -1.2236744165420532,
      "step": 630
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.827470069914772e-06,
      "logits/chosen": -4.29737663269043,
      "logits/rejected": -4.263535499572754,
      "logps/chosen": -376.9189453125,
      "logps/rejected": -404.60247802734375,
      "loss": 0.6551,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -1.0662126541137695,
      "rewards/margins": 0.14261041581630707,
      "rewards/rejected": -1.2088232040405273,
      "step": 640
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.7580779535868675e-06,
      "logits/chosen": -4.269147872924805,
      "logits/rejected": -4.276608467102051,
      "logps/chosen": -366.36492919921875,
      "logps/rejected": -390.6036376953125,
      "loss": 0.6583,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.9485187530517578,
      "rewards/margins": 0.15207722783088684,
      "rewards/rejected": -1.1005959510803223,
      "step": 650
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.688484237617129e-06,
      "logits/chosen": -4.208004474639893,
      "logits/rejected": -4.182279586791992,
      "logps/chosen": -362.94586181640625,
      "logps/rejected": -393.9739990234375,
      "loss": 0.6322,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.8923367261886597,
      "rewards/margins": 0.20138171315193176,
      "rewards/rejected": -1.0937182903289795,
      "step": 660
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.6187432856891585e-06,
      "logits/chosen": -4.1537322998046875,
      "logits/rejected": -4.151703357696533,
      "logps/chosen": -372.31201171875,
      "logps/rejected": -405.5547180175781,
      "loss": 0.6489,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.953785240650177,
      "rewards/margins": 0.15488819777965546,
      "rewards/rejected": -1.108673334121704,
      "step": 670
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.548909576501096e-06,
      "logits/chosen": -4.182751655578613,
      "logits/rejected": -4.181895732879639,
      "logps/chosen": -372.65435791015625,
      "logps/rejected": -398.80902099609375,
      "loss": 0.6442,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.9745942950248718,
      "rewards/margins": 0.160622701048851,
      "rewards/rejected": -1.1352169513702393,
      "step": 680
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.4790376612091503e-06,
      "logits/chosen": -4.2613630294799805,
      "logits/rejected": -4.227023124694824,
      "logps/chosen": -391.2955017089844,
      "logps/rejected": -419.54656982421875,
      "loss": 0.6233,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.0887335538864136,
      "rewards/margins": 0.2128380537033081,
      "rewards/rejected": -1.3015714883804321,
      "step": 690
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.40918212081453e-06,
      "logits/chosen": -4.308588981628418,
      "logits/rejected": -4.264666557312012,
      "logps/chosen": -360.5702819824219,
      "logps/rejected": -409.14422607421875,
      "loss": 0.5974,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.0383028984069824,
      "rewards/margins": 0.29866236448287964,
      "rewards/rejected": -1.3369653224945068,
      "step": 700
    },
    {
      "epoch": 0.56,
      "eval_logits/chosen": -4.447990417480469,
      "eval_logits/rejected": -4.436967849731445,
      "eval_logps/chosen": -322.94854736328125,
      "eval_logps/rejected": -338.0312194824219,
      "eval_loss": 0.6777762174606323,
      "eval_rewards/accuracies": 0.5653333067893982,
      "eval_rewards/chosen": -1.1454858779907227,
      "eval_rewards/margins": 0.09772318601608276,
      "eval_rewards/rejected": -1.2432090044021606,
      "eval_runtime": 1566.0394,
      "eval_samples_per_second": 1.912,
      "eval_steps_per_second": 0.239,
      "step": 700
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.3393975235270654e-06,
      "logits/chosen": -4.278247833251953,
      "logits/rejected": -4.263747692108154,
      "logps/chosen": -394.7868347167969,
      "logps/rejected": -433.32373046875,
      "loss": 0.6411,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -1.1327521800994873,
      "rewards/margins": 0.22177498042583466,
      "rewards/rejected": -1.354527235031128,
      "step": 710
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2697383821388153e-06,
      "logits/chosen": -4.283724308013916,
      "logits/rejected": -4.2910637855529785,
      "logps/chosen": -383.4498596191406,
      "logps/rejected": -406.142578125,
      "loss": 0.6479,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.1030470132827759,
      "rewards/margins": 0.19455528259277344,
      "rewards/rejected": -1.2976024150848389,
      "step": 720
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2002591114409657e-06,
      "logits/chosen": -4.243215560913086,
      "logits/rejected": -4.230997562408447,
      "logps/chosen": -392.0690612792969,
      "logps/rejected": -419.01092529296875,
      "loss": 0.6517,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": -1.1523319482803345,
      "rewards/margins": 0.17104671895503998,
      "rewards/rejected": -1.323378562927246,
      "step": 730
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.131013985717285e-06,
      "logits/chosen": -4.2963361740112305,
      "logits/rejected": -4.248631477355957,
      "logps/chosen": -402.02130126953125,
      "logps/rejected": -451.04400634765625,
      "loss": 0.6211,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.197863221168518,
      "rewards/margins": 0.26271212100982666,
      "rewards/rejected": -1.4605753421783447,
      "step": 740
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.062057096347338e-06,
      "logits/chosen": -4.2642927169799805,
      "logits/rejected": -4.245224952697754,
      "logps/chosen": -386.35394287109375,
      "logps/rejected": -399.6516418457031,
      "loss": 0.6611,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.1031478643417358,
      "rewards/margins": 0.12747173011302948,
      "rewards/rejected": -1.2306196689605713,
      "step": 750
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9934423095525733e-06,
      "logits/chosen": -4.144771575927734,
      "logits/rejected": -4.146181583404541,
      "logps/chosen": -381.7077941894531,
      "logps/rejected": -401.769287109375,
      "loss": 0.6555,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.9921267628669739,
      "rewards/margins": 0.19879314303398132,
      "rewards/rejected": -1.190920114517212,
      "step": 760
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9252232243182986e-06,
      "logits/chosen": -4.229475498199463,
      "logits/rejected": -4.177742958068848,
      "logps/chosen": -345.7065734863281,
      "logps/rejected": -388.74462890625,
      "loss": 0.6144,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.8882685899734497,
      "rewards/margins": 0.26380977034568787,
      "rewards/rejected": -1.15207839012146,
      "step": 770
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.8574531305244043e-06,
      "logits/chosen": -4.0229105949401855,
      "logits/rejected": -4.011281490325928,
      "logps/chosen": -381.92205810546875,
      "logps/rejected": -418.5514221191406,
      "loss": 0.6197,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.013808250427246,
      "rewards/margins": 0.23891642689704895,
      "rewards/rejected": -1.2527246475219727,
      "step": 780
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.7901849673175559e-06,
      "logits/chosen": -4.090008735656738,
      "logits/rejected": -4.056280612945557,
      "logps/chosen": -375.6288146972656,
      "logps/rejected": -409.71490478515625,
      "loss": 0.634,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.9446262121200562,
      "rewards/margins": 0.2019636631011963,
      "rewards/rejected": -1.1465898752212524,
      "step": 790
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7234712817573555e-06,
      "logits/chosen": -4.081742763519287,
      "logits/rejected": -4.069535255432129,
      "logps/chosen": -408.01580810546875,
      "logps/rejected": -428.19830322265625,
      "loss": 0.6508,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.1142655611038208,
      "rewards/margins": 0.18682199716567993,
      "rewards/rejected": -1.301087498664856,
      "step": 800
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -4.271145343780518,
      "eval_logits/rejected": -4.261835098266602,
      "eval_logps/chosen": -318.4167785644531,
      "eval_logps/rejected": -333.9435119628906,
      "eval_loss": 0.6748104691505432,
      "eval_rewards/accuracies": 0.5649999976158142,
      "eval_rewards/chosen": -1.1001687049865723,
      "eval_rewards/margins": 0.10216327011585236,
      "eval_rewards/rejected": -1.2023320198059082,
      "eval_runtime": 1566.0255,
      "eval_samples_per_second": 1.912,
      "eval_steps_per_second": 0.239,
      "step": 800
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.6573641877687936e-06,
      "logits/chosen": -4.110816478729248,
      "logits/rejected": -4.093571662902832,
      "logps/chosen": -381.05877685546875,
      "logps/rejected": -423.17364501953125,
      "loss": 0.6202,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.0917795896530151,
      "rewards/margins": 0.21704904735088348,
      "rewards/rejected": -1.308828592300415,
      "step": 810
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.591915325433034e-06,
      "logits/chosen": -4.177929878234863,
      "logits/rejected": -4.188526630401611,
      "logps/chosen": -370.15655517578125,
      "logps/rejected": -401.153564453125,
      "loss": 0.6176,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.020612359046936,
      "rewards/margins": 0.26860466599464417,
      "rewards/rejected": -1.2892169952392578,
      "step": 820
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5271758206483664e-06,
      "logits/chosen": -4.208674430847168,
      "logits/rejected": -4.197338104248047,
      "logps/chosen": -399.302490234375,
      "logps/rejected": -428.2151794433594,
      "loss": 0.6565,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.2150713205337524,
      "rewards/margins": 0.18397213518619537,
      "rewards/rejected": -1.3990434408187866,
      "step": 830
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4631962451927966e-06,
      "logits/chosen": -4.116250514984131,
      "logits/rejected": -4.103257656097412,
      "logps/chosen": -396.60467529296875,
      "logps/rejected": -433.5016174316406,
      "loss": 0.6155,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.1897716522216797,
      "rewards/margins": 0.23897738754749298,
      "rewards/rejected": -1.4287490844726562,
      "step": 840
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.4000265772195032e-06,
      "logits/chosen": -4.289558410644531,
      "logits/rejected": -4.233942985534668,
      "logps/chosen": -391.7837219238281,
      "logps/rejected": -433.24261474609375,
      "loss": 0.6144,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.1339555978775024,
      "rewards/margins": 0.25826430320739746,
      "rewards/rejected": -1.3922197818756104,
      "step": 850
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3377161622160077e-06,
      "logits/chosen": -4.226014137268066,
      "logits/rejected": -4.22386360168457,
      "logps/chosen": -392.7188415527344,
      "logps/rejected": -428.2996520996094,
      "loss": 0.6116,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.1736845970153809,
      "rewards/margins": 0.23641009628772736,
      "rewards/rejected": -1.4100948572158813,
      "step": 860
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.276313674457553e-06,
      "logits/chosen": -4.321101188659668,
      "logits/rejected": -4.316857814788818,
      "logps/chosen": -376.7432556152344,
      "logps/rejected": -420.355224609375,
      "loss": 0.6062,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.1783860921859741,
      "rewards/margins": 0.2604502737522125,
      "rewards/rejected": -1.4388363361358643,
      "step": 870
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.2158670789848095e-06,
      "logits/chosen": -4.357269287109375,
      "logits/rejected": -4.3427653312683105,
      "logps/chosen": -412.49603271484375,
      "logps/rejected": -446.0122985839844,
      "loss": 0.6207,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.3438339233398438,
      "rewards/margins": 0.2622073292732239,
      "rewards/rejected": -1.6060411930084229,
      "step": 880
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.1564235941356016e-06,
      "logits/chosen": -4.453230381011963,
      "logits/rejected": -4.401095390319824,
      "logps/chosen": -398.6358337402344,
      "logps/rejected": -450.500244140625,
      "loss": 0.6262,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.3652451038360596,
      "rewards/margins": 0.3177439570426941,
      "rewards/rejected": -1.6829888820648193,
      "step": 890
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.0980296546599254e-06,
      "logits/chosen": -4.331755638122559,
      "logits/rejected": -4.3203840255737305,
      "logps/chosen": -422.264404296875,
      "logps/rejected": -445.89739990234375,
      "loss": 0.6746,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -1.3867082595825195,
      "rewards/margins": 0.28304582834243774,
      "rewards/rejected": -1.6697540283203125,
      "step": 900
    },
    {
      "epoch": 0.72,
      "eval_logits/chosen": -4.577242374420166,
      "eval_logits/rejected": -4.566162586212158,
      "eval_logps/chosen": -341.29400634765625,
      "eval_logps/rejected": -358.15576171875,
      "eval_loss": 0.6757350564002991,
      "eval_rewards/accuracies": 0.5686666369438171,
      "eval_rewards/chosen": -1.3289406299591064,
      "eval_rewards/margins": 0.11551385372877121,
      "eval_rewards/rejected": -1.4444544315338135,
      "eval_runtime": 1566.2371,
      "eval_samples_per_second": 1.912,
      "eval_steps_per_second": 0.239,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.040730875447083e-06,
      "logits/chosen": -4.31623649597168,
      "logits/rejected": -4.329934597015381,
      "logps/chosen": -407.89544677734375,
      "logps/rejected": -431.8106384277344,
      "loss": 0.6508,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.282503604888916,
      "rewards/margins": 0.16822777688503265,
      "rewards/rejected": -1.4507312774658203,
      "step": 910
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.845720158932414e-07,
      "logits/chosen": -4.317078590393066,
      "logits/rejected": -4.328789710998535,
      "logps/chosen": -384.6829833984375,
      "logps/rejected": -404.79107666015625,
      "loss": 0.6351,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -1.2274483442306519,
      "rewards/margins": 0.19847750663757324,
      "rewards/rejected": -1.4259259700775146,
      "step": 920
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.295969449372796e-07,
      "logits/chosen": -4.379042625427246,
      "logits/rejected": -4.360438346862793,
      "logps/chosen": -385.20245361328125,
      "logps/rejected": -416.4458923339844,
      "loss": 0.6352,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -1.1640335321426392,
      "rewards/margins": 0.19592450559139252,
      "rewards/rejected": -1.3599579334259033,
      "step": 930
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.758486067922176e-07,
      "logits/chosen": -4.383969306945801,
      "logits/rejected": -4.344205379486084,
      "logps/chosen": -384.3717346191406,
      "logps/rejected": -427.7298889160156,
      "loss": 0.6135,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -1.207880973815918,
      "rewards/margins": 0.2520582675933838,
      "rewards/rejected": -1.4599391222000122,
      "step": 940
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.233689873990006e-07,
      "logits/chosen": -4.391127109527588,
      "logits/rejected": -4.357823848724365,
      "logps/chosen": -389.40570068359375,
      "logps/rejected": -443.3841247558594,
      "loss": 0.6165,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.2565414905548096,
      "rewards/margins": 0.3726215362548828,
      "rewards/rejected": -1.6291630268096924,
      "step": 950
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.721990816288555e-07,
      "logits/chosen": -4.3561787605285645,
      "logits/rejected": -4.322731971740723,
      "logps/chosen": -363.4510803222656,
      "logps/rejected": -400.2040710449219,
      "loss": 0.6128,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.1432923078536987,
      "rewards/margins": 0.24396805465221405,
      "rewards/rejected": -1.3872603178024292,
      "step": 960
    },
    {
      "epoch": 0.78,
      "learning_rate": 7.223788612598148e-07,
      "logits/chosen": -4.374851703643799,
      "logits/rejected": -4.3536810874938965,
      "logps/chosen": -397.66680908203125,
      "logps/rejected": -428.77227783203125,
      "loss": 0.639,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.2013370990753174,
      "rewards/margins": 0.15685971081256866,
      "rewards/rejected": -1.3581968545913696,
      "step": 970
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.73947243752448e-07,
      "logits/chosen": -4.431571006774902,
      "logits/rejected": -4.440985679626465,
      "logps/chosen": -415.3916931152344,
      "logps/rejected": -431.7449645996094,
      "loss": 0.656,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.3690643310546875,
      "rewards/margins": 0.22862248122692108,
      "rewards/rejected": -1.5976868867874146,
      "step": 980
    },
    {
      "epoch": 0.79,
      "learning_rate": 6.269420618491759e-07,
      "logits/chosen": -4.3968610763549805,
      "logits/rejected": -4.365904808044434,
      "logps/chosen": -384.7822265625,
      "logps/rejected": -417.7525939941406,
      "loss": 0.6459,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.2827244997024536,
      "rewards/margins": 0.17925159633159637,
      "rewards/rejected": -1.461976170539856,
      "step": 990
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.814000340209267e-07,
      "logits/chosen": -4.392641067504883,
      "logits/rejected": -4.334468841552734,
      "logps/chosen": -398.47662353515625,
      "logps/rejected": -445.6675720214844,
      "loss": 0.6151,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.2670553922653198,
      "rewards/margins": 0.2884874939918518,
      "rewards/rejected": -1.5555428266525269,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "eval_logits/chosen": -4.618439197540283,
      "eval_logits/rejected": -4.607028484344482,
      "eval_logps/chosen": -343.98931884765625,
      "eval_logps/rejected": -361.1741943359375,
      "eval_loss": 0.6755098700523376,
      "eval_rewards/accuracies": 0.5690000057220459,
      "eval_rewards/chosen": -1.3558939695358276,
      "eval_rewards/margins": 0.11874517053365707,
      "eval_rewards/rejected": -1.4746389389038086,
      "eval_runtime": 1565.397,
      "eval_samples_per_second": 1.913,
      "eval_steps_per_second": 0.24,
      "step": 1000
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.373567357842111e-07,
      "logits/chosen": -4.363351821899414,
      "logits/rejected": -4.336891174316406,
      "logps/chosen": -393.9075012207031,
      "logps/rejected": -432.1922302246094,
      "loss": 0.6011,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -1.2202444076538086,
      "rewards/margins": 0.2966518700122833,
      "rewards/rejected": -1.516896367073059,
      "step": 1010
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.948465719110226e-07,
      "logits/chosen": -4.430235862731934,
      "logits/rejected": -4.40949010848999,
      "logps/chosen": -379.41375732421875,
      "logps/rejected": -404.0538635253906,
      "loss": 0.6481,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -1.1881725788116455,
      "rewards/margins": 0.17870518565177917,
      "rewards/rejected": -1.366877794265747,
      "step": 1020
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.539027495532766e-07,
      "logits/chosen": -4.403011322021484,
      "logits/rejected": -4.427884578704834,
      "logps/chosen": -381.40582275390625,
      "logps/rejected": -415.64044189453125,
      "loss": 0.6201,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.2115557193756104,
      "rewards/margins": 0.2422032654285431,
      "rewards/rejected": -1.4537590742111206,
      "step": 1030
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.14557252302783e-07,
      "logits/chosen": -4.416018962860107,
      "logits/rejected": -4.378045082092285,
      "logps/chosen": -401.25286865234375,
      "logps/rejected": -434.80523681640625,
      "loss": 0.6795,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -1.350279450416565,
      "rewards/margins": 0.18791106343269348,
      "rewards/rejected": -1.538190484046936,
      "step": 1040
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.7684081520700884e-07,
      "logits/chosen": -4.323966026306152,
      "logits/rejected": -4.319721221923828,
      "logps/chosen": -411.38812255859375,
      "logps/rejected": -426.6982421875,
      "loss": 0.6426,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.2739022970199585,
      "rewards/margins": 0.20139846205711365,
      "rewards/rejected": -1.475300908088684,
      "step": 1050
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.407829007601507e-07,
      "logits/chosen": -4.3614420890808105,
      "logits/rejected": -4.317663192749023,
      "logps/chosen": -395.0092468261719,
      "logps/rejected": -438.6324157714844,
      "loss": 0.6132,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -1.2276242971420288,
      "rewards/margins": 0.28230905532836914,
      "rewards/rejected": -1.5099334716796875,
      "step": 1060
    },
    {
      "epoch": 0.86,
      "learning_rate": 3.064116758882724e-07,
      "logits/chosen": -4.330022811889648,
      "logits/rejected": -4.274947643280029,
      "logps/chosen": -406.9677429199219,
      "logps/rejected": -459.18048095703125,
      "loss": 0.6063,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.304404616355896,
      "rewards/margins": 0.33513855934143066,
      "rewards/rejected": -1.6395431756973267,
      "step": 1070
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.737539899464908e-07,
      "logits/chosen": -4.385619163513184,
      "logits/rejected": -4.378711700439453,
      "logps/chosen": -370.54156494140625,
      "logps/rejected": -411.8634338378906,
      "loss": 0.6197,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.1883795261383057,
      "rewards/margins": 0.2932683825492859,
      "rewards/rejected": -1.4816478490829468,
      "step": 1080
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.4283535374538645e-07,
      "logits/chosen": -4.3373942375183105,
      "logits/rejected": -4.339685916900635,
      "logps/chosen": -400.1960754394531,
      "logps/rejected": -430.27557373046875,
      "loss": 0.6378,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -1.2183153629302979,
      "rewards/margins": 0.2266024649143219,
      "rewards/rejected": -1.444917917251587,
      "step": 1090
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.1367991962303298e-07,
      "logits/chosen": -4.3288164138793945,
      "logits/rejected": -4.31368350982666,
      "logps/chosen": -382.65814208984375,
      "logps/rejected": -405.8753967285156,
      "loss": 0.6837,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -1.2913353443145752,
      "rewards/margins": 0.10410255193710327,
      "rewards/rejected": -1.3954378366470337,
      "step": 1100
    },
    {
      "epoch": 0.88,
      "eval_logits/chosen": -4.571716785430908,
      "eval_logits/rejected": -4.5607380867004395,
      "eval_logps/chosen": -340.8576354980469,
      "eval_logps/rejected": -358.08392333984375,
      "eval_loss": 0.6748068928718567,
      "eval_rewards/accuracies": 0.5709999799728394,
      "eval_rewards/chosen": -1.3245766162872314,
      "eval_rewards/margins": 0.11915931850671768,
      "eval_rewards/rejected": -1.4437358379364014,
      "eval_runtime": 1563.7394,
      "eval_samples_per_second": 1.915,
      "eval_steps_per_second": 0.24,
      "step": 1100
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.8631046257820278e-07,
      "logits/chosen": -4.340999603271484,
      "logits/rejected": -4.353137969970703,
      "logps/chosen": -399.4720153808594,
      "logps/rejected": -427.0414123535156,
      "loss": 0.6277,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.2465407848358154,
      "rewards/margins": 0.23380056023597717,
      "rewards/rejected": -1.4803413152694702,
      "step": 1110
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.6074836247950143e-07,
      "logits/chosen": -4.328183174133301,
      "logits/rejected": -4.327004432678223,
      "logps/chosen": -406.1539306640625,
      "logps/rejected": -430.02044677734375,
      "loss": 0.6483,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.2271990776062012,
      "rewards/margins": 0.17618249356746674,
      "rewards/rejected": -1.4033817052841187,
      "step": 1120
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.370135873643097e-07,
      "logits/chosen": -4.33071231842041,
      "logits/rejected": -4.364874839782715,
      "logps/chosen": -403.37628173828125,
      "logps/rejected": -413.484375,
      "loss": 0.6427,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -1.2655575275421143,
      "rewards/margins": 0.19985264539718628,
      "rewards/rejected": -1.4654099941253662,
      "step": 1130
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.1512467784059372e-07,
      "logits/chosen": -4.4061384201049805,
      "logits/rejected": -4.369538307189941,
      "logps/chosen": -367.4423522949219,
      "logps/rejected": -402.65911865234375,
      "loss": 0.6186,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -1.1879181861877441,
      "rewards/margins": 0.27168983221054077,
      "rewards/rejected": -1.4596080780029297,
      "step": 1140
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.509873260376251e-08,
      "logits/chosen": -4.3491435050964355,
      "logits/rejected": -4.2928547859191895,
      "logps/chosen": -394.6898498535156,
      "logps/rejected": -461.10845947265625,
      "loss": 0.583,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.2356455326080322,
      "rewards/margins": 0.3976030647754669,
      "rewards/rejected": -1.6332485675811768,
      "step": 1150
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.695139507988559e-08,
      "logits/chosen": -4.359416961669922,
      "logits/rejected": -4.381959915161133,
      "logps/chosen": -409.65643310546875,
      "logps/rejected": -434.98675537109375,
      "loss": 0.6521,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -1.2608633041381836,
      "rewards/margins": 0.1726626455783844,
      "rewards/rejected": -1.4335259199142456,
      "step": 1160
    },
    {
      "epoch": 0.94,
      "learning_rate": 6.069684120570684e-08,
      "logits/chosen": -4.381880283355713,
      "logits/rejected": -4.29121208190918,
      "logps/chosen": -395.4755554199219,
      "logps/rejected": -438.13427734375,
      "loss": 0.6418,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.3113107681274414,
      "rewards/margins": 0.23519746959209442,
      "rewards/rejected": -1.5465081930160522,
      "step": 1170
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.634776835499871e-08,
      "logits/chosen": -4.314273357391357,
      "logits/rejected": -4.296536445617676,
      "logps/chosen": -382.25958251953125,
      "logps/rejected": -419.88897705078125,
      "loss": 0.6354,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -1.2596309185028076,
      "rewards/margins": 0.24562938511371613,
      "rewards/rejected": -1.5052604675292969,
      "step": 1180
    },
    {
      "epoch": 0.95,
      "learning_rate": 3.3915385419908964e-08,
      "logits/chosen": -4.292089939117432,
      "logits/rejected": -4.3260178565979,
      "logps/chosen": -391.3450927734375,
      "logps/rejected": -419.9061584472656,
      "loss": 0.6332,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -1.2074798345565796,
      "rewards/margins": 0.2409057915210724,
      "rewards/rejected": -1.4483857154846191,
      "step": 1190
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.3409404055043938e-08,
      "logits/chosen": -4.403561592102051,
      "logits/rejected": -4.3759870529174805,
      "logps/chosen": -400.16522216796875,
      "logps/rejected": -427.40606689453125,
      "loss": 0.6539,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -1.288109302520752,
      "rewards/margins": 0.17616990208625793,
      "rewards/rejected": -1.4642791748046875,
      "step": 1200
    },
    {
      "epoch": 0.96,
      "eval_logits/chosen": -4.559507846832275,
      "eval_logits/rejected": -4.548335075378418,
      "eval_logps/chosen": -340.22308349609375,
      "eval_logps/rejected": -357.4035949707031,
      "eval_loss": 0.6746300458908081,
      "eval_rewards/accuracies": 0.5730000138282776,
      "eval_rewards/chosen": -1.318231225013733,
      "eval_rewards/margins": 0.11870139837265015,
      "eval_rewards/rejected": -1.4369325637817383,
      "eval_runtime": 1563.0275,
      "eval_samples_per_second": 1.916,
      "eval_steps_per_second": 0.24,
      "step": 1200
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4838031091134186e-08,
      "logits/chosen": -4.341926574707031,
      "logits/rejected": -4.2704925537109375,
      "logps/chosen": -373.17681884765625,
      "logps/rejected": -431.8099060058594,
      "loss": 0.6065,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.2158408164978027,
      "rewards/margins": 0.33925992250442505,
      "rewards/rejected": -1.555100679397583,
      "step": 1210
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.207962124201774e-09,
      "logits/chosen": -4.371233940124512,
      "logits/rejected": -4.343363285064697,
      "logps/chosen": -401.8610534667969,
      "logps/rejected": -437.0784606933594,
      "loss": 0.6005,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -1.1877615451812744,
      "rewards/margins": 0.29727452993392944,
      "rewards/rejected": -1.4850361347198486,
      "step": 1220
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.5243762852441023e-09,
      "logits/chosen": -4.321781158447266,
      "logits/rejected": -4.277449131011963,
      "logps/chosen": -392.8436279296875,
      "logps/rejected": -430.36279296875,
      "loss": 0.6401,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.2518620491027832,
      "rewards/margins": 0.2195155918598175,
      "rewards/rejected": -1.4713777303695679,
      "step": 1230
    },
    {
      "epoch": 0.99,
      "learning_rate": 7.909321945129278e-10,
      "logits/chosen": -4.296388626098633,
      "logits/rejected": -4.257707118988037,
      "logps/chosen": -408.13409423828125,
      "logps/rejected": -451.42462158203125,
      "loss": 0.6106,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -1.2207772731781006,
      "rewards/margins": 0.29706281423568726,
      "rewards/rejected": -1.5178401470184326,
      "step": 1240
    },
    {
      "epoch": 1.0,
      "step": 1249,
      "total_flos": 0.0,
      "train_loss": 0.6508825262419409,
      "train_runtime": 40390.1729,
      "train_samples_per_second": 0.495,
      "train_steps_per_second": 0.031
    }
  ],
  "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
}