{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 84,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011904761904761904,
      "grad_norm": 11.529353141784668,
      "learning_rate": 2e-05,
      "loss": 1.914004921913147,
      "mean_token_accuracy": 0.5769918859004974,
      "num_tokens": 29788.0,
      "step": 1
    },
    {
      "epoch": 0.023809523809523808,
      "grad_norm": 14.16106128692627,
      "learning_rate": 1.999922292480975e-05,
      "loss": 1.83119797706604,
      "mean_token_accuracy": 0.5946373492479324,
      "num_tokens": 58751.0,
      "step": 2
    },
    {
      "epoch": 0.03571428571428571,
      "grad_norm": 9.59426498413086,
      "learning_rate": 1.9996891820008165e-05,
      "loss": 1.8222196102142334,
      "mean_token_accuracy": 0.6167362779378891,
      "num_tokens": 87987.0,
      "step": 3
    },
    {
      "epoch": 0.047619047619047616,
      "grad_norm": 5.701657772064209,
      "learning_rate": 1.9993007047883988e-05,
      "loss": 1.4995148181915283,
      "mean_token_accuracy": 0.6572036817669868,
      "num_tokens": 116847.0,
      "step": 4
    },
    {
      "epoch": 0.05952380952380952,
      "grad_norm": 4.609629154205322,
      "learning_rate": 1.9987569212189224e-05,
      "loss": 1.3815604448318481,
      "mean_token_accuracy": 0.6702215820550919,
      "num_tokens": 146324.0,
      "step": 5
    },
    {
      "epoch": 0.07142857142857142,
      "grad_norm": 3.5629072189331055,
      "learning_rate": 1.9980579158045322e-05,
      "loss": 1.34389066696167,
      "mean_token_accuracy": 0.6741709485650063,
      "num_tokens": 174803.0,
      "step": 6
    },
    {
      "epoch": 0.08333333333333333,
      "grad_norm": 3.0423154830932617,
      "learning_rate": 1.9972037971811802e-05,
      "loss": 1.275808572769165,
      "mean_token_accuracy": 0.6867651119828224,
      "num_tokens": 203369.0,
      "step": 7
    },
    {
      "epoch": 0.09523809523809523,
      "grad_norm": 2.9051754474639893,
      "learning_rate": 1.9961946980917457e-05,
      "loss": 1.1854076385498047,
      "mean_token_accuracy": 0.7050081267952919,
      "num_tokens": 232982.0,
      "step": 8
    },
    {
      "epoch": 0.10714285714285714,
      "grad_norm": 2.5599188804626465,
      "learning_rate": 1.9950307753654016e-05,
      "loss": 1.2208622694015503,
      "mean_token_accuracy": 0.6956963464617729,
      "num_tokens": 261286.0,
      "step": 9
    },
    {
      "epoch": 0.11904761904761904,
      "grad_norm": 2.9105069637298584,
      "learning_rate": 1.9937122098932428e-05,
      "loss": 1.1587761640548706,
      "mean_token_accuracy": 0.7048550546169281,
      "num_tokens": 290542.0,
      "step": 10
    },
    {
      "epoch": 0.13095238095238096,
      "grad_norm": 2.3259518146514893,
      "learning_rate": 1.9922392066001724e-05,
      "loss": 1.0967768430709839,
      "mean_token_accuracy": 0.7165249660611153,
      "num_tokens": 320601.0,
      "step": 11
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 2.5200023651123047,
      "learning_rate": 1.9906119944130527e-05,
      "loss": 1.0702282190322876,
      "mean_token_accuracy": 0.7203697860240936,
      "num_tokens": 350202.0,
      "step": 12
    },
    {
      "epoch": 0.15476190476190477,
      "grad_norm": 2.1715333461761475,
      "learning_rate": 1.9888308262251286e-05,
      "loss": 1.0621695518493652,
      "mean_token_accuracy": 0.7225284278392792,
      "num_tokens": 379636.0,
      "step": 13
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": 2.610119342803955,
      "learning_rate": 1.9868959788567213e-05,
      "loss": 1.0720399618148804,
      "mean_token_accuracy": 0.7212414294481277,
      "num_tokens": 406951.0,
      "step": 14
    },
    {
      "epoch": 0.17857142857142858,
      "grad_norm": 2.6167454719543457,
      "learning_rate": 1.9848077530122083e-05,
      "loss": 1.0404419898986816,
      "mean_token_accuracy": 0.7257834300398827,
      "num_tokens": 435217.0,
      "step": 15
    },
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 2.4558708667755127,
      "learning_rate": 1.9825664732332886e-05,
      "loss": 1.1281776428222656,
      "mean_token_accuracy": 0.707080751657486,
      "num_tokens": 464425.0,
      "step": 16
    },
    {
      "epoch": 0.20238095238095238,
      "grad_norm": 2.375856876373291,
      "learning_rate": 1.9801724878485438e-05,
      "loss": 1.0459752082824707,
      "mean_token_accuracy": 0.7244894206523895,
      "num_tokens": 492566.0,
      "step": 17
    },
    {
      "epoch": 0.21428571428571427,
      "grad_norm": 2.2667899131774902,
      "learning_rate": 1.977626168919305e-05,
      "loss": 0.9850942492485046,
      "mean_token_accuracy": 0.7457425370812416,
      "num_tokens": 522039.0,
      "step": 18
    },
    {
      "epoch": 0.2261904761904762,
      "grad_norm": 2.1298699378967285,
      "learning_rate": 1.9749279121818235e-05,
      "loss": 1.0323350429534912,
      "mean_token_accuracy": 0.73154316842556,
      "num_tokens": 551228.0,
      "step": 19
    },
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 1.9165962934494019,
      "learning_rate": 1.9720781369857747e-05,
      "loss": 0.994964599609375,
      "mean_token_accuracy": 0.7338539212942123,
      "num_tokens": 579813.0,
      "step": 20
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.1407601833343506,
      "learning_rate": 1.969077286229078e-05,
      "loss": 1.0337525606155396,
      "mean_token_accuracy": 0.7247511297464371,
      "num_tokens": 609054.0,
      "step": 21
    },
    {
      "epoch": 0.2619047619047619,
      "grad_norm": 2.2245829105377197,
      "learning_rate": 1.9659258262890683e-05,
      "loss": 0.9542327523231506,
      "mean_token_accuracy": 0.7403879463672638,
      "num_tokens": 638361.0,
      "step": 22
    },
    {
      "epoch": 0.27380952380952384,
      "grad_norm": 2.094996213912964,
      "learning_rate": 1.962624246950012e-05,
      "loss": 0.9909444451332092,
      "mean_token_accuracy": 0.7375892624258995,
      "num_tokens": 667030.0,
      "step": 23
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 2.194431781768799,
      "learning_rate": 1.9591730613269878e-05,
      "loss": 1.0016077756881714,
      "mean_token_accuracy": 0.7301788702607155,
      "num_tokens": 695951.0,
      "step": 24
    },
    {
      "epoch": 0.2976190476190476,
      "grad_norm": 2.199652671813965,
      "learning_rate": 1.955572805786141e-05,
      "loss": 0.956201434135437,
      "mean_token_accuracy": 0.7408725842833519,
      "num_tokens": 725170.0,
      "step": 25
    },
    {
      "epoch": 0.30952380952380953,
      "grad_norm": 2.0795445442199707,
      "learning_rate": 1.9518240398613226e-05,
      "loss": 0.9647458791732788,
      "mean_token_accuracy": 0.7420727759599686,
      "num_tokens": 754824.0,
      "step": 26
    },
    {
      "epoch": 0.32142857142857145,
      "grad_norm": 1.9338266849517822,
      "learning_rate": 1.947927346167132e-05,
      "loss": 0.992991030216217,
      "mean_token_accuracy": 0.731611043214798,
      "num_tokens": 784104.0,
      "step": 27
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 1.9213852882385254,
      "learning_rate": 1.9438833303083677e-05,
      "loss": 0.9164495468139648,
      "mean_token_accuracy": 0.7498249039053917,
      "num_tokens": 813113.0,
      "step": 28
    },
    {
      "epoch": 0.34523809523809523,
      "grad_norm": 1.8558398485183716,
      "learning_rate": 1.9396926207859085e-05,
      "loss": 1.006327509880066,
      "mean_token_accuracy": 0.7329650148749352,
      "num_tokens": 842641.0,
      "step": 29
    },
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 2.002457857131958,
      "learning_rate": 1.935355868899034e-05,
      "loss": 0.9234254956245422,
      "mean_token_accuracy": 0.7504403591156006,
      "num_tokens": 870913.0,
      "step": 30
    },
    {
      "epoch": 0.36904761904761907,
      "grad_norm": 1.9878448247909546,
      "learning_rate": 1.9308737486442045e-05,
      "loss": 0.9712892174720764,
      "mean_token_accuracy": 0.7372883632779121,
      "num_tokens": 899828.0,
      "step": 31
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 1.9190301895141602,
      "learning_rate": 1.926246956610309e-05,
      "loss": 0.8859049081802368,
      "mean_token_accuracy": 0.7609693109989166,
      "num_tokens": 928435.0,
      "step": 32
    },
    {
      "epoch": 0.39285714285714285,
      "grad_norm": 1.81462562084198,
      "learning_rate": 1.921476211870408e-05,
      "loss": 0.9458773136138916,
      "mean_token_accuracy": 0.7440405786037445,
      "num_tokens": 957815.0,
      "step": 33
    },
    {
      "epoch": 0.40476190476190477,
      "grad_norm": 1.9442874193191528,
      "learning_rate": 1.9165622558699763e-05,
      "loss": 0.9346192479133606,
      "mean_token_accuracy": 0.7506085485219955,
      "num_tokens": 986586.0,
      "step": 34
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 1.7711018323898315,
      "learning_rate": 1.9115058523116734e-05,
      "loss": 0.9014881253242493,
      "mean_token_accuracy": 0.7530476674437523,
      "num_tokens": 1015838.0,
      "step": 35
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 1.801094651222229,
      "learning_rate": 1.9063077870366504e-05,
      "loss": 0.9483398199081421,
      "mean_token_accuracy": 0.7421744614839554,
      "num_tokens": 1045470.0,
      "step": 36
    },
    {
      "epoch": 0.44047619047619047,
      "grad_norm": 1.792850375175476,
      "learning_rate": 1.900968867902419e-05,
      "loss": 0.9111062288284302,
      "mean_token_accuracy": 0.752615176141262,
      "num_tokens": 1073211.0,
      "step": 37
    },
    {
      "epoch": 0.4523809523809524,
      "grad_norm": 1.8647798299789429,
      "learning_rate": 1.895489924657301e-05,
      "loss": 0.8721588253974915,
      "mean_token_accuracy": 0.7621221914887428,
      "num_tokens": 1102374.0,
      "step": 38
    },
    {
      "epoch": 0.4642857142857143,
      "grad_norm": 1.6813071966171265,
      "learning_rate": 1.8898718088114688e-05,
      "loss": 0.8819330930709839,
      "mean_token_accuracy": 0.7566907480359077,
      "num_tokens": 1131350.0,
      "step": 39
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 1.6970678567886353,
      "learning_rate": 1.8841153935046098e-05,
      "loss": 0.8742820620536804,
      "mean_token_accuracy": 0.760765790939331,
      "num_tokens": 1160238.0,
      "step": 40
    },
    {
      "epoch": 0.4880952380952381,
      "grad_norm": 1.7569754123687744,
      "learning_rate": 1.8782215733702286e-05,
      "loss": 0.8570113182067871,
      "mean_token_accuracy": 0.7613107413053513,
      "num_tokens": 1189331.0,
      "step": 41
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.782785177230835,
      "learning_rate": 1.8721912643966055e-05,
      "loss": 0.9163686037063599,
      "mean_token_accuracy": 0.7505383789539337,
      "num_tokens": 1217448.0,
      "step": 42
    },
    {
      "epoch": 0.5119047619047619,
      "grad_norm": 1.6833268404006958,
      "learning_rate": 1.866025403784439e-05,
      "loss": 0.8989430665969849,
      "mean_token_accuracy": 0.7541684806346893,
      "num_tokens": 1247253.0,
      "step": 43
    },
    {
      "epoch": 0.5238095238095238,
      "grad_norm": 1.6909689903259277,
      "learning_rate": 1.8597249498011906e-05,
      "loss": 0.8409149050712585,
      "mean_token_accuracy": 0.7660783603787422,
      "num_tokens": 1275646.0,
      "step": 44
    },
    {
      "epoch": 0.5357142857142857,
      "grad_norm": 1.8078761100769043,
      "learning_rate": 1.8532908816321557e-05,
      "loss": 0.9054464101791382,
      "mean_token_accuracy": 0.7482258230447769,
      "num_tokens": 1304469.0,
      "step": 45
    },
    {
      "epoch": 0.5476190476190477,
      "grad_norm": 1.6278305053710938,
      "learning_rate": 1.8467241992282842e-05,
      "loss": 0.9137347936630249,
      "mean_token_accuracy": 0.746067963540554,
      "num_tokens": 1333024.0,
      "step": 46
    },
    {
      "epoch": 0.5595238095238095,
      "grad_norm": 1.695748209953308,
      "learning_rate": 1.8400259231507716e-05,
      "loss": 0.8484127521514893,
      "mean_token_accuracy": 0.7649081125855446,
      "num_tokens": 1361283.0,
      "step": 47
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 1.7918659448623657,
      "learning_rate": 1.833197094412449e-05,
      "loss": 0.8585721850395203,
      "mean_token_accuracy": 0.7619551867246628,
      "num_tokens": 1389627.0,
      "step": 48
    },
    {
      "epoch": 0.5833333333333334,
      "grad_norm": 1.6668473482131958,
      "learning_rate": 1.826238774315995e-05,
      "loss": 0.8342987895011902,
      "mean_token_accuracy": 0.7647690623998642,
      "num_tokens": 1418097.0,
      "step": 49
    },
    {
      "epoch": 0.5952380952380952,
      "grad_norm": 1.7863699197769165,
      "learning_rate": 1.819152044288992e-05,
      "loss": 0.9001743793487549,
      "mean_token_accuracy": 0.750338077545166,
      "num_tokens": 1446015.0,
      "step": 50
    },
    {
      "epoch": 0.6071428571428571,
      "grad_norm": 1.71127450466156,
      "learning_rate": 1.811938005715857e-05,
      "loss": 0.8569691181182861,
      "mean_token_accuracy": 0.759001724421978,
      "num_tokens": 1474484.0,
      "step": 51
    },
    {
      "epoch": 0.6190476190476191,
      "grad_norm": 1.7378276586532593,
      "learning_rate": 1.8045977797666685e-05,
      "loss": 0.9362456798553467,
      "mean_token_accuracy": 0.7411532327532768,
      "num_tokens": 1502138.0,
      "step": 52
    },
    {
      "epoch": 0.6309523809523809,
      "grad_norm": 1.6661640405654907,
      "learning_rate": 1.7971325072229227e-05,
      "loss": 0.9059334397315979,
      "mean_token_accuracy": 0.7488799020648003,
      "num_tokens": 1531714.0,
      "step": 53
    },
    {
      "epoch": 0.6428571428571429,
      "grad_norm": 1.6824532747268677,
      "learning_rate": 1.7895433483002356e-05,
      "loss": 0.9190329313278198,
      "mean_token_accuracy": 0.7526443004608154,
      "num_tokens": 1559571.0,
      "step": 54
    },
    {
      "epoch": 0.6547619047619048,
      "grad_norm": 1.676474928855896,
      "learning_rate": 1.78183148246803e-05,
      "loss": 0.8781886100769043,
      "mean_token_accuracy": 0.7565796673297882,
      "num_tokens": 1588454.0,
      "step": 55
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 1.6373399496078491,
      "learning_rate": 1.7739981082662275e-05,
      "loss": 0.8755627870559692,
      "mean_token_accuracy": 0.7579803913831711,
      "num_tokens": 1618492.0,
      "step": 56
    },
    {
      "epoch": 0.6785714285714286,
      "grad_norm": 1.6949313879013062,
      "learning_rate": 1.766044443118978e-05,
      "loss": 0.8839727640151978,
      "mean_token_accuracy": 0.7543486133217812,
      "num_tokens": 1646795.0,
      "step": 57
    },
    {
      "epoch": 0.6904761904761905,
      "grad_norm": 1.690186858177185,
      "learning_rate": 1.757971723145453e-05,
      "loss": 0.8561735153198242,
      "mean_token_accuracy": 0.7641277685761452,
      "num_tokens": 1675431.0,
      "step": 58
    },
    {
      "epoch": 0.7023809523809523,
      "grad_norm": 1.6995313167572021,
      "learning_rate": 1.7497812029677344e-05,
      "loss": 0.8403012156486511,
      "mean_token_accuracy": 0.7634185254573822,
      "num_tokens": 1702955.0,
      "step": 59
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 1.6898642778396606,
      "learning_rate": 1.741474155515827e-05,
      "loss": 0.8668173551559448,
      "mean_token_accuracy": 0.75686926394701,
      "num_tokens": 1732138.0,
      "step": 60
    },
    {
      "epoch": 0.7261904761904762,
      "grad_norm": 1.6695095300674438,
      "learning_rate": 1.7330518718298263e-05,
      "loss": 0.8799718022346497,
      "mean_token_accuracy": 0.7577245235443115,
      "num_tokens": 1761445.0,
      "step": 61
    },
    {
      "epoch": 0.7380952380952381,
      "grad_norm": 1.6258364915847778,
      "learning_rate": 1.7245156608592727e-05,
      "loss": 0.8522840142250061,
      "mean_token_accuracy": 0.7693983167409897,
      "num_tokens": 1791067.0,
      "step": 62
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.7827128171920776,
      "learning_rate": 1.7158668492597186e-05,
      "loss": 0.9087042212486267,
      "mean_token_accuracy": 0.7470668181777,
      "num_tokens": 1818881.0,
      "step": 63
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 1.677538514137268,
      "learning_rate": 1.7071067811865477e-05,
      "loss": 0.8644732236862183,
      "mean_token_accuracy": 0.7561531737446785,
      "num_tokens": 1847402.0,
      "step": 64
    },
    {
      "epoch": 0.7738095238095238,
      "grad_norm": 1.6414068937301636,
      "learning_rate": 1.698236818086073e-05,
      "loss": 0.8923105001449585,
      "mean_token_accuracy": 0.7542363926768303,
      "num_tokens": 1876445.0,
      "step": 65
    },
    {
      "epoch": 0.7857142857142857,
      "grad_norm": 1.5977094173431396,
      "learning_rate": 1.689258338483947e-05,
      "loss": 0.8339341282844543,
      "mean_token_accuracy": 0.7679391428828239,
      "num_tokens": 1905528.0,
      "step": 66
    },
    {
      "epoch": 0.7976190476190477,
      "grad_norm": 1.5793612003326416,
      "learning_rate": 1.6801727377709195e-05,
      "loss": 0.836018443107605,
      "mean_token_accuracy": 0.7610953524708748,
      "num_tokens": 1933975.0,
      "step": 67
    },
    {
      "epoch": 0.8095238095238095,
      "grad_norm": 1.6230100393295288,
      "learning_rate": 1.67098142798597e-05,
      "loss": 0.8169003129005432,
      "mean_token_accuracy": 0.7718362584710121,
      "num_tokens": 1962702.0,
      "step": 68
    },
    {
      "epoch": 0.8214285714285714,
      "grad_norm": 1.6503868103027344,
      "learning_rate": 1.6616858375968596e-05,
      "loss": 0.851630449295044,
      "mean_token_accuracy": 0.7629296258091927,
      "num_tokens": 1991343.0,
      "step": 69
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 1.5629642009735107,
      "learning_rate": 1.6522874112781213e-05,
      "loss": 0.8485338091850281,
      "mean_token_accuracy": 0.7656759098172188,
      "num_tokens": 2020170.0,
      "step": 70
    },
    {
      "epoch": 0.8452380952380952,
      "grad_norm": 1.6736105680465698,
      "learning_rate": 1.6427876096865394e-05,
      "loss": 0.8417783379554749,
      "mean_token_accuracy": 0.7630589976906776,
      "num_tokens": 2050423.0,
      "step": 71
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 1.595051884651184,
      "learning_rate": 1.6331879092341402e-05,
      "loss": 0.8609828352928162,
      "mean_token_accuracy": 0.7579501941800117,
      "num_tokens": 2079518.0,
      "step": 72
    },
    {
      "epoch": 0.8690476190476191,
      "grad_norm": 1.5752646923065186,
      "learning_rate": 1.6234898018587336e-05,
      "loss": 0.790068507194519,
      "mean_token_accuracy": 0.7756508365273476,
      "num_tokens": 2109234.0,
      "step": 73
    },
    {
      "epoch": 0.8809523809523809,
      "grad_norm": 1.6872154474258423,
      "learning_rate": 1.6136947947920477e-05,
      "loss": 0.8501731157302856,
      "mean_token_accuracy": 0.7624671682715416,
      "num_tokens": 2138017.0,
      "step": 74
    },
    {
      "epoch": 0.8928571428571429,
      "grad_norm": 1.553470253944397,
      "learning_rate": 1.6038044103254775e-05,
      "loss": 0.7876694202423096,
      "mean_token_accuracy": 0.7766945138573647,
      "num_tokens": 2167971.0,
      "step": 75
    },
    {
      "epoch": 0.9047619047619048,
      "grad_norm": 1.5356166362762451,
      "learning_rate": 1.5938201855735017e-05,
      "loss": 0.864933431148529,
      "mean_token_accuracy": 0.758280374109745,
      "num_tokens": 2196376.0,
      "step": 76
    },
    {
      "epoch": 0.9166666666666666,
      "grad_norm": 1.5473679304122925,
      "learning_rate": 1.5837436722347902e-05,
      "loss": 0.7725358009338379,
      "mean_token_accuracy": 0.783719077706337,
      "num_tokens": 2225589.0,
      "step": 77
    },
    {
      "epoch": 0.9285714285714286,
      "grad_norm": 1.595386028289795,
      "learning_rate": 1.573576436351046e-05,
      "loss": 0.8284271955490112,
      "mean_token_accuracy": 0.7694306820631027,
      "num_tokens": 2254878.0,
      "step": 78
    },
    {
      "epoch": 0.9404761904761905,
      "grad_norm": 1.5001078844070435,
      "learning_rate": 1.563320058063622e-05,
      "loss": 0.7898463010787964,
      "mean_token_accuracy": 0.7755530849099159,
      "num_tokens": 2284143.0,
      "step": 79
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 1.5622379779815674,
      "learning_rate": 1.5529761313679396e-05,
      "loss": 0.8229933977127075,
      "mean_token_accuracy": 0.7678125724196434,
      "num_tokens": 2312406.0,
      "step": 80
    },
    {
      "epoch": 0.9642857142857143,
      "grad_norm": 1.6046059131622314,
      "learning_rate": 1.5425462638657597e-05,
      "loss": 0.8793967366218567,
      "mean_token_accuracy": 0.7589751705527306,
      "num_tokens": 2342080.0,
      "step": 81
    },
    {
      "epoch": 0.9761904761904762,
      "grad_norm": 1.5159045457839966,
      "learning_rate": 1.5320320765153367e-05,
      "loss": 0.788962721824646,
      "mean_token_accuracy": 0.7769099771976471,
      "num_tokens": 2370973.0,
      "step": 82
    },
    {
      "epoch": 0.9880952380952381,
      "grad_norm": 1.5396736860275269,
      "learning_rate": 1.5214352033794981e-05,
      "loss": 0.8291895389556885,
      "mean_token_accuracy": 0.7664894834160805,
      "num_tokens": 2399858.0,
      "step": 83
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.656612753868103,
      "learning_rate": 1.5107572913716859e-05,
      "loss": 0.8644437193870544,
      "mean_token_accuracy": 0.7582353055477142,
      "num_tokens": 2427247.0,
      "step": 84
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8236262798309326,
      "eval_mean_token_accuracy": 0.7665031782786051,
      "eval_num_tokens": 2427247.0,
      "eval_runtime": 30.1941,
      "eval_samples_per_second": 4.968,
      "eval_steps_per_second": 4.968,
      "step": 84
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 252,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.302169712014336e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}