{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9978902953586498,
  "eval_steps": 500,
  "global_step": 473,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010548523206751054,
      "grad_norm": 35.3808479309082,
      "learning_rate": 7.904761904761904e-06,
      "loss": 3.2969,
      "mean_token_accuracy": 0.5261552810668946,
      "num_tokens": 165661.0,
      "step": 5
    },
    {
      "epoch": 0.02109704641350211,
      "grad_norm": 2.607041120529175,
      "learning_rate": 1.7785714285714282e-05,
      "loss": 0.7292,
      "mean_token_accuracy": 0.7687696695327759,
      "num_tokens": 330871.0,
      "step": 10
    },
    {
      "epoch": 0.03164556962025317,
      "grad_norm": 0.45053747296333313,
      "learning_rate": 2.7666666666666667e-05,
      "loss": 0.2006,
      "mean_token_accuracy": 0.8882377624511719,
      "num_tokens": 497005.0,
      "step": 15
    },
    {
      "epoch": 0.04219409282700422,
      "grad_norm": 0.35783207416534424,
      "learning_rate": 3.7547619047619045e-05,
      "loss": 0.1758,
      "mean_token_accuracy": 0.8900018930435181,
      "num_tokens": 665392.0,
      "step": 20
    },
    {
      "epoch": 0.052742616033755275,
      "grad_norm": 0.1681824028491974,
      "learning_rate": 4.7428571428571427e-05,
      "loss": 0.1807,
      "mean_token_accuracy": 0.8822107434272766,
      "num_tokens": 829859.0,
      "step": 25
    },
    {
      "epoch": 0.06329113924050633,
      "grad_norm": 0.19937705993652344,
      "learning_rate": 5.730952380952381e-05,
      "loss": 0.1797,
      "mean_token_accuracy": 0.8809401750564575,
      "num_tokens": 997784.0,
      "step": 30
    },
    {
      "epoch": 0.07383966244725738,
      "grad_norm": 0.16169631481170654,
      "learning_rate": 6.719047619047618e-05,
      "loss": 0.1825,
      "mean_token_accuracy": 0.8777167439460755,
      "num_tokens": 1165766.0,
      "step": 35
    },
    {
      "epoch": 0.08438818565400844,
      "grad_norm": 0.061657316982746124,
      "learning_rate": 6.915604092749773e-05,
      "loss": 0.1772,
      "mean_token_accuracy": 0.8926807641983032,
      "num_tokens": 1332443.0,
      "step": 40
    },
    {
      "epoch": 0.0949367088607595,
      "grad_norm": 0.08636178821325302,
      "learning_rate": 6.911288878267097e-05,
      "loss": 0.1752,
      "mean_token_accuracy": 0.8955834746360779,
      "num_tokens": 1498172.0,
      "step": 45
    },
    {
      "epoch": 0.10548523206751055,
      "grad_norm": 0.09138909727334976,
      "learning_rate": 6.903660132263935e-05,
      "loss": 0.1739,
      "mean_token_accuracy": 0.8818711161613464,
      "num_tokens": 1664579.0,
      "step": 50
    },
    {
      "epoch": 0.1160337552742616,
      "grad_norm": 0.11239784210920334,
      "learning_rate": 6.892727620758576e-05,
      "loss": 0.18,
      "mean_token_accuracy": 0.8861486673355102,
      "num_tokens": 1830686.0,
      "step": 55
    },
    {
      "epoch": 0.12658227848101267,
      "grad_norm": 0.1455163061618805,
      "learning_rate": 6.878505339118957e-05,
      "loss": 0.1811,
      "mean_token_accuracy": 0.8851905465126038,
      "num_tokens": 1998499.0,
      "step": 60
    },
    {
      "epoch": 0.1371308016877637,
      "grad_norm": 0.09706641733646393,
      "learning_rate": 6.861011494146364e-05,
      "loss": 0.1781,
      "mean_token_accuracy": 0.8888298153877259,
      "num_tokens": 2164511.0,
      "step": 65
    },
    {
      "epoch": 0.14767932489451477,
      "grad_norm": 0.04830857366323471,
      "learning_rate": 6.840268480767798e-05,
      "loss": 0.1763,
      "mean_token_accuracy": 0.8890364646911622,
      "num_tokens": 2332502.0,
      "step": 70
    },
    {
      "epoch": 0.15822784810126583,
      "grad_norm": 0.06256038695573807,
      "learning_rate": 6.816302853366881e-05,
      "loss": 0.1752,
      "mean_token_accuracy": 0.8831726312637329,
      "num_tokens": 2500395.0,
      "step": 75
    },
    {
      "epoch": 0.16877637130801687,
      "grad_norm": 0.11051741242408752,
      "learning_rate": 6.789145291789983e-05,
      "loss": 0.1827,
      "mean_token_accuracy": 0.8779117345809937,
      "num_tokens": 2666867.0,
      "step": 80
    },
    {
      "epoch": 0.17932489451476794,
      "grad_norm": 0.10065566003322601,
      "learning_rate": 6.7588305620711e-05,
      "loss": 0.1741,
      "mean_token_accuracy": 0.8870775103569031,
      "num_tokens": 2833858.0,
      "step": 85
    },
    {
      "epoch": 0.189873417721519,
      "grad_norm": 0.07600655406713486,
      "learning_rate": 6.725397471925754e-05,
      "loss": 0.1759,
      "mean_token_accuracy": 0.8873356342315674,
      "num_tokens": 3002010.0,
      "step": 90
    },
    {
      "epoch": 0.20042194092827004,
      "grad_norm": 0.06981377303600311,
      "learning_rate": 6.6888888210709e-05,
      "loss": 0.1785,
      "mean_token_accuracy": 0.8861026048660279,
      "num_tokens": 3168178.0,
      "step": 95
    },
    {
      "epoch": 0.2109704641350211,
      "grad_norm": 0.06852678209543228,
      "learning_rate": 6.649351346434416e-05,
      "loss": 0.1785,
      "mean_token_accuracy": 0.8875195026397705,
      "num_tokens": 3336666.0,
      "step": 100
    },
    {
      "epoch": 0.22151898734177214,
      "grad_norm": 0.05367989093065262,
      "learning_rate": 6.606835662324347e-05,
      "loss": 0.1769,
      "mean_token_accuracy": 0.8868713140487671,
      "num_tokens": 3501349.0,
      "step": 105
    },
    {
      "epoch": 0.2320675105485232,
      "grad_norm": 0.03881872817873955,
      "learning_rate": 6.561396195634468e-05,
      "loss": 0.1734,
      "mean_token_accuracy": 0.8922244787216187,
      "num_tokens": 3667080.0,
      "step": 110
    },
    {
      "epoch": 0.24261603375527427,
      "grad_norm": 0.056945957243442535,
      "learning_rate": 6.513091116169134e-05,
      "loss": 0.1743,
      "mean_token_accuracy": 0.8882174730300904,
      "num_tokens": 3832403.0,
      "step": 115
    },
    {
      "epoch": 0.25316455696202533,
      "grad_norm": 0.06432785838842392,
      "learning_rate": 6.461982262176606e-05,
      "loss": 0.1737,
      "mean_token_accuracy": 0.886626124382019,
      "num_tokens": 3997743.0,
      "step": 120
    },
    {
      "epoch": 0.26371308016877637,
      "grad_norm": 0.06393298506736755,
      "learning_rate": 6.408135061186171e-05,
      "loss": 0.1785,
      "mean_token_accuracy": 0.8915424346923828,
      "num_tokens": 4164839.0,
      "step": 125
    },
    {
      "epoch": 0.2742616033755274,
      "grad_norm": 0.09579639136791229,
      "learning_rate": 6.351618446250413e-05,
      "loss": 0.182,
      "mean_token_accuracy": 0.883327579498291,
      "num_tokens": 4331140.0,
      "step": 130
    },
    {
      "epoch": 0.2848101265822785,
      "grad_norm": 0.08194269984960556,
      "learning_rate": 6.292504767699852e-05,
      "loss": 0.1708,
      "mean_token_accuracy": 0.8873592734336853,
      "num_tokens": 4499701.0,
      "step": 135
    },
    {
      "epoch": 0.29535864978902954,
      "grad_norm": 0.07841971516609192,
      "learning_rate": 6.230869700522922e-05,
      "loss": 0.1748,
      "mean_token_accuracy": 0.8862677931785583,
      "num_tokens": 4663587.0,
      "step": 140
    },
    {
      "epoch": 0.3059071729957806,
      "grad_norm": 0.07796195149421692,
      "learning_rate": 6.166792147489844e-05,
      "loss": 0.1754,
      "mean_token_accuracy": 0.8887560129165649,
      "num_tokens": 4830134.0,
      "step": 145
    },
    {
      "epoch": 0.31645569620253167,
      "grad_norm": 0.0773933008313179,
      "learning_rate": 6.100354138144424e-05,
      "loss": 0.1773,
      "mean_token_accuracy": 0.8887858510017395,
      "num_tokens": 4997630.0,
      "step": 150
    },
    {
      "epoch": 0.3270042194092827,
      "grad_norm": 0.06370408833026886,
      "learning_rate": 6.0316407237930975e-05,
      "loss": 0.1765,
      "mean_token_accuracy": 0.8919444918632508,
      "num_tokens": 5166237.0,
      "step": 155
    },
    {
      "epoch": 0.33755274261603374,
      "grad_norm": 0.06905322521924973,
      "learning_rate": 5.9607398686256e-05,
      "loss": 0.1768,
      "mean_token_accuracy": 0.8846820235252381,
      "num_tokens": 5332877.0,
      "step": 160
    },
    {
      "epoch": 0.34810126582278483,
      "grad_norm": 0.06343474239110947,
      "learning_rate": 5.887742337106709e-05,
      "loss": 0.1831,
      "mean_token_accuracy": 0.8806776642799378,
      "num_tokens": 5499208.0,
      "step": 165
    },
    {
      "epoch": 0.35864978902953587,
      "grad_norm": 0.05995175614953041,
      "learning_rate": 5.812741577783165e-05,
      "loss": 0.1796,
      "mean_token_accuracy": 0.8807573795318604,
      "num_tokens": 5664741.0,
      "step": 170
    },
    {
      "epoch": 0.3691983122362869,
      "grad_norm": 0.04446087405085564,
      "learning_rate": 5.735833603654556e-05,
      "loss": 0.1691,
      "mean_token_accuracy": 0.892852509021759,
      "num_tokens": 5834918.0,
      "step": 175
    },
    {
      "epoch": 0.379746835443038,
      "grad_norm": 0.0720236673951149,
      "learning_rate": 5.657116869261271e-05,
      "loss": 0.1739,
      "mean_token_accuracy": 0.894946813583374,
      "num_tokens": 6001503.0,
      "step": 180
    },
    {
      "epoch": 0.39029535864978904,
      "grad_norm": 0.05321015045046806,
      "learning_rate": 5.576692144646905e-05,
      "loss": 0.1794,
      "mean_token_accuracy": 0.8884535312652588,
      "num_tokens": 6164580.0,
      "step": 185
    },
    {
      "epoch": 0.4008438818565401,
      "grad_norm": 0.07356124371290207,
      "learning_rate": 5.494662386356439e-05,
      "loss": 0.1761,
      "mean_token_accuracy": 0.884639322757721,
      "num_tokens": 6332797.0,
      "step": 190
    },
    {
      "epoch": 0.41139240506329117,
      "grad_norm": 0.07975731045007706,
      "learning_rate": 5.411132605635359e-05,
      "loss": 0.1811,
      "mean_token_accuracy": 0.884186589717865,
      "num_tokens": 6501501.0,
      "step": 195
    },
    {
      "epoch": 0.4219409282700422,
      "grad_norm": 0.10269442200660706,
      "learning_rate": 5.32620973399841e-05,
      "loss": 0.1756,
      "mean_token_accuracy": 0.8871823906898498,
      "num_tokens": 6667819.0,
      "step": 200
    },
    {
      "epoch": 0.43248945147679324,
      "grad_norm": 0.09555568546056747,
      "learning_rate": 5.24000248634012e-05,
      "loss": 0.1816,
      "mean_token_accuracy": 0.8835095286369323,
      "num_tokens": 6832488.0,
      "step": 205
    },
    {
      "epoch": 0.4430379746835443,
      "grad_norm": 0.10852270573377609,
      "learning_rate": 5.1526212217622945e-05,
      "loss": 0.1791,
      "mean_token_accuracy": 0.8864225625991822,
      "num_tokens": 6997947.0,
      "step": 210
    },
    {
      "epoch": 0.45358649789029537,
      "grad_norm": 0.07506605237722397,
      "learning_rate": 5.064177802296666e-05,
      "loss": 0.1773,
      "mean_token_accuracy": 0.8899453759193421,
      "num_tokens": 7166746.0,
      "step": 215
    },
    {
      "epoch": 0.4641350210970464,
      "grad_norm": 0.054569222033023834,
      "learning_rate": 4.974785449703566e-05,
      "loss": 0.1769,
      "mean_token_accuracy": 0.8855153083801269,
      "num_tokens": 7332336.0,
      "step": 220
    },
    {
      "epoch": 0.47468354430379744,
      "grad_norm": 0.06732705235481262,
      "learning_rate": 4.8845586005299005e-05,
      "loss": 0.178,
      "mean_token_accuracy": 0.884535837173462,
      "num_tokens": 7497795.0,
      "step": 225
    },
    {
      "epoch": 0.48523206751054854,
      "grad_norm": 0.14481715857982635,
      "learning_rate": 4.793612759612026e-05,
      "loss": 0.1755,
      "mean_token_accuracy": 0.8890446305274964,
      "num_tokens": 7663986.0,
      "step": 230
    },
    {
      "epoch": 0.4957805907172996,
      "grad_norm": 0.1290741264820099,
      "learning_rate": 4.7020643522110336e-05,
      "loss": 0.1788,
      "mean_token_accuracy": 0.8863259434700013,
      "num_tokens": 7827636.0,
      "step": 235
    },
    {
      "epoch": 0.5063291139240507,
      "grad_norm": 0.052244700491428375,
      "learning_rate": 4.6100305749697454e-05,
      "loss": 0.1739,
      "mean_token_accuracy": 0.887740409374237,
      "num_tokens": 7992516.0,
      "step": 240
    },
    {
      "epoch": 0.5168776371308017,
      "grad_norm": 0.14647985994815826,
      "learning_rate": 4.517629245882217e-05,
      "loss": 0.1743,
      "mean_token_accuracy": 0.8877281546592712,
      "num_tokens": 8160127.0,
      "step": 245
    },
    {
      "epoch": 0.5274261603375527,
      "grad_norm": 0.06328880041837692,
      "learning_rate": 4.4249786534678096e-05,
      "loss": 0.1777,
      "mean_token_accuracy": 0.8938660621643066,
      "num_tokens": 8327284.0,
      "step": 250
    },
    {
      "epoch": 0.5379746835443038,
      "grad_norm": 0.06663276255130768,
      "learning_rate": 4.332197405342916e-05,
      "loss": 0.1775,
      "mean_token_accuracy": 0.8839066505432129,
      "num_tokens": 8496122.0,
      "step": 255
    },
    {
      "epoch": 0.5485232067510548,
      "grad_norm": 0.040306951850652695,
      "learning_rate": 4.239404276384192e-05,
      "loss": 0.1725,
      "mean_token_accuracy": 0.897387170791626,
      "num_tokens": 8662324.0,
      "step": 260
    },
    {
      "epoch": 0.5590717299578059,
      "grad_norm": 0.13088253140449524,
      "learning_rate": 4.146718056677666e-05,
      "loss": 0.1789,
      "mean_token_accuracy": 0.8823062062263489,
      "num_tokens": 8826605.0,
      "step": 265
    },
    {
      "epoch": 0.569620253164557,
      "grad_norm": 0.08320745825767517,
      "learning_rate": 4.0542573994483785e-05,
      "loss": 0.1722,
      "mean_token_accuracy": 0.8971045613288879,
      "num_tokens": 8993205.0,
      "step": 270
    },
    {
      "epoch": 0.580168776371308,
      "grad_norm": 0.057557735592126846,
      "learning_rate": 3.962140669165228e-05,
      "loss": 0.1726,
      "mean_token_accuracy": 0.8942949891090393,
      "num_tokens": 9162034.0,
      "step": 275
    },
    {
      "epoch": 0.5907172995780591,
      "grad_norm": 0.0712743028998375,
      "learning_rate": 3.8704857900154624e-05,
      "loss": 0.1768,
      "mean_token_accuracy": 0.8844752430915832,
      "num_tokens": 9328749.0,
      "step": 280
    },
    {
      "epoch": 0.6012658227848101,
      "grad_norm": 0.07155586779117584,
      "learning_rate": 3.77941009494282e-05,
      "loss": 0.1686,
      "mean_token_accuracy": 0.8906964659690857,
      "num_tokens": 9494827.0,
      "step": 285
    },
    {
      "epoch": 0.6118143459915611,
      "grad_norm": 0.12148404866456985,
      "learning_rate": 3.689030175442532e-05,
      "loss": 0.1719,
      "mean_token_accuracy": 0.8948503375053406,
      "num_tokens": 9661052.0,
      "step": 290
    },
    {
      "epoch": 0.6223628691983122,
      "grad_norm": 0.058917637914419174,
      "learning_rate": 3.599461732305523e-05,
      "loss": 0.1715,
      "mean_token_accuracy": 0.8935618162155151,
      "num_tokens": 9829664.0,
      "step": 295
    },
    {
      "epoch": 0.6329113924050633,
      "grad_norm": 0.052568018436431885,
      "learning_rate": 3.510819427502835e-05,
      "loss": 0.1813,
      "mean_token_accuracy": 0.8813806295394897,
      "num_tokens": 9991047.0,
      "step": 300
    },
    {
      "epoch": 0.6434599156118144,
      "grad_norm": 0.039732322096824646,
      "learning_rate": 3.423216737399921e-05,
      "loss": 0.179,
      "mean_token_accuracy": 0.882714855670929,
      "num_tokens": 10156808.0,
      "step": 305
    },
    {
      "epoch": 0.6540084388185654,
      "grad_norm": 0.08463432639837265,
      "learning_rate": 3.336765807488699e-05,
      "loss": 0.1748,
      "mean_token_accuracy": 0.888425087928772,
      "num_tokens": 10321114.0,
      "step": 310
    },
    {
      "epoch": 0.6645569620253164,
      "grad_norm": 0.11946665495634079,
      "learning_rate": 3.251577308823336e-05,
      "loss": 0.1737,
      "mean_token_accuracy": 0.8930611729621887,
      "num_tokens": 10488983.0,
      "step": 315
    },
    {
      "epoch": 0.6751054852320675,
      "grad_norm": 0.06477019935846329,
      "learning_rate": 3.167760296343553e-05,
      "loss": 0.1798,
      "mean_token_accuracy": 0.8830307364463806,
      "num_tokens": 10654617.0,
      "step": 320
    },
    {
      "epoch": 0.6856540084388185,
      "grad_norm": 0.10155260562896729,
      "learning_rate": 3.0854220692668106e-05,
      "loss": 0.1798,
      "mean_token_accuracy": 0.8801281929016114,
      "num_tokens": 10817430.0,
      "step": 325
    },
    {
      "epoch": 0.6962025316455697,
      "grad_norm": 0.07864528894424438,
      "learning_rate": 3.0046680337281083e-05,
      "loss": 0.169,
      "mean_token_accuracy": 0.8891819953918457,
      "num_tokens": 10983444.0,
      "step": 330
    },
    {
      "epoch": 0.7067510548523207,
      "grad_norm": 0.08378790318965912,
      "learning_rate": 2.9256015678432208e-05,
      "loss": 0.1731,
      "mean_token_accuracy": 0.8935487747192383,
      "num_tokens": 11147338.0,
      "step": 335
    },
    {
      "epoch": 0.7172995780590717,
      "grad_norm": 0.04839184507727623,
      "learning_rate": 2.848323889368137e-05,
      "loss": 0.1704,
      "mean_token_accuracy": 0.8966975808143616,
      "num_tokens": 11314767.0,
      "step": 340
    },
    {
      "epoch": 0.7278481012658228,
      "grad_norm": 0.06478141993284225,
      "learning_rate": 2.7729339261241013e-05,
      "loss": 0.172,
      "mean_token_accuracy": 0.8940050840377808,
      "num_tokens": 11482091.0,
      "step": 345
    },
    {
      "epoch": 0.7383966244725738,
      "grad_norm": 0.06571247428655624,
      "learning_rate": 2.69952818935413e-05,
      "loss": 0.1665,
      "mean_token_accuracy": 0.891292929649353,
      "num_tokens": 11647898.0,
      "step": 350
    },
    {
      "epoch": 0.7489451476793249,
      "grad_norm": 0.11665584146976471,
      "learning_rate": 2.6282006501731506e-05,
      "loss": 0.1759,
      "mean_token_accuracy": 0.8836230754852294,
      "num_tokens": 11813898.0,
      "step": 355
    },
    {
      "epoch": 0.759493670886076,
      "grad_norm": 0.052924949675798416,
      "learning_rate": 2.5590426192699053e-05,
      "loss": 0.1649,
      "mean_token_accuracy": 0.8995304107666016,
      "num_tokens": 11981645.0,
      "step": 360
    },
    {
      "epoch": 0.770042194092827,
      "grad_norm": 0.07151809334754944,
      "learning_rate": 2.4921426300146175e-05,
      "loss": 0.1707,
      "mean_token_accuracy": 0.893867826461792,
      "num_tokens": 12150084.0,
      "step": 365
    },
    {
      "epoch": 0.7805907172995781,
      "grad_norm": 0.06855164468288422,
      "learning_rate": 2.4275863251220893e-05,
      "loss": 0.1747,
      "mean_token_accuracy": 0.8869495272636414,
      "num_tokens": 12318806.0,
      "step": 370
    },
    {
      "epoch": 0.7911392405063291,
      "grad_norm": 0.08721784502267838,
      "learning_rate": 2.3654563470152875e-05,
      "loss": 0.1772,
      "mean_token_accuracy": 0.8878260612487793,
      "num_tokens": 12486999.0,
      "step": 375
    },
    {
      "epoch": 0.8016877637130801,
      "grad_norm": 0.08103222399950027,
      "learning_rate": 2.3058322320297915e-05,
      "loss": 0.1793,
      "mean_token_accuracy": 0.8905072569847107,
      "num_tokens": 12653461.0,
      "step": 380
    },
    {
      "epoch": 0.8122362869198312,
      "grad_norm": 0.06833580136299133,
      "learning_rate": 2.2487903085945242e-05,
      "loss": 0.1732,
      "mean_token_accuracy": 0.888560950756073,
      "num_tokens": 12819702.0,
      "step": 385
    },
    {
      "epoch": 0.8227848101265823,
      "grad_norm": 0.038626912981271744,
      "learning_rate": 2.1944035995191313e-05,
      "loss": 0.1711,
      "mean_token_accuracy": 0.8971449255943298,
      "num_tokens": 12985584.0,
      "step": 390
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 0.04700719192624092,
      "learning_rate": 2.142741728513063e-05,
      "loss": 0.171,
      "mean_token_accuracy": 0.8911542415618896,
      "num_tokens": 13153362.0,
      "step": 395
    },
    {
      "epoch": 0.8438818565400844,
      "grad_norm": 0.04693188890814781,
      "learning_rate": 2.0938708310560715e-05,
      "loss": 0.1713,
      "mean_token_accuracy": 0.8950452446937561,
      "num_tokens": 13318640.0,
      "step": 400
    },
    {
      "epoch": 0.8544303797468354,
      "grad_norm": 0.06232467293739319,
      "learning_rate": 2.0478534697341822e-05,
      "loss": 0.1713,
      "mean_token_accuracy": 0.8938077449798584,
      "num_tokens": 13485225.0,
      "step": 405
    },
    {
      "epoch": 0.8649789029535865,
      "grad_norm": 0.055438246577978134,
      "learning_rate": 2.004748554149557e-05,
      "loss": 0.1744,
      "mean_token_accuracy": 0.8867436647415161,
      "num_tokens": 13650422.0,
      "step": 410
    },
    {
      "epoch": 0.8755274261603375,
      "grad_norm": 0.05567546933889389,
      "learning_rate": 1.9646112655067517e-05,
      "loss": 0.1743,
      "mean_token_accuracy": 0.8970485091209411,
      "num_tokens": 13819123.0,
      "step": 415
    },
    {
      "epoch": 0.8860759493670886,
      "grad_norm": 0.055763162672519684,
      "learning_rate": 1.927492985971929e-05,
      "loss": 0.1728,
      "mean_token_accuracy": 0.894312059879303,
      "num_tokens": 13988886.0,
      "step": 420
    },
    {
      "epoch": 0.8966244725738397,
      "grad_norm": 0.048897162079811096,
      "learning_rate": 1.8934412328954416e-05,
      "loss": 0.1734,
      "mean_token_accuracy": 0.8896780252456665,
      "num_tokens": 14154966.0,
      "step": 425
    },
    {
      "epoch": 0.9071729957805907,
      "grad_norm": 0.048425015062093735,
      "learning_rate": 1.8624995979820078e-05,
      "loss": 0.1704,
      "mean_token_accuracy": 0.8915553092956543,
      "num_tokens": 14321918.0,
      "step": 430
    },
    {
      "epoch": 0.9177215189873418,
      "grad_norm": 0.0759928971529007,
      "learning_rate": 1.8347076914863353e-05,
      "loss": 0.175,
      "mean_token_accuracy": 0.9015556693077087,
      "num_tokens": 14490294.0,
      "step": 435
    },
    {
      "epoch": 0.9282700421940928,
      "grad_norm": 0.03949494659900665,
      "learning_rate": 1.810101091505645e-05,
      "loss": 0.1689,
      "mean_token_accuracy": 0.9001257538795471,
      "num_tokens": 14657195.0,
      "step": 440
    },
    {
      "epoch": 0.9388185654008439,
      "grad_norm": 0.05298343673348427,
      "learning_rate": 1.788711298434e-05,
      "loss": 0.1721,
      "mean_token_accuracy": 0.8874968409538269,
      "num_tokens": 14823893.0,
      "step": 445
    },
    {
      "epoch": 0.9493670886075949,
      "grad_norm": 0.05960085988044739,
      "learning_rate": 1.7705656946367505e-05,
      "loss": 0.169,
      "mean_token_accuracy": 0.8921529054641724,
      "num_tokens": 14990386.0,
      "step": 450
    },
    {
      "epoch": 0.959915611814346,
      "grad_norm": 0.09964242577552795,
      "learning_rate": 1.7556875093967134e-05,
      "loss": 0.1709,
      "mean_token_accuracy": 0.8904130220413208,
      "num_tokens": 15157796.0,
      "step": 455
    },
    {
      "epoch": 0.9704641350210971,
      "grad_norm": 0.10357833653688431,
      "learning_rate": 1.744095789176968e-05,
      "loss": 0.1736,
      "mean_token_accuracy": 0.8919926643371582,
      "num_tokens": 15322983.0,
      "step": 460
    },
    {
      "epoch": 0.9810126582278481,
      "grad_norm": 0.04409240186214447,
      "learning_rate": 1.735805373238324e-05,
      "loss": 0.1722,
      "mean_token_accuracy": 0.8972856163978576,
      "num_tokens": 15489671.0,
      "step": 465
    },
    {
      "epoch": 0.9915611814345991,
      "grad_norm": 0.05001898482441902,
      "learning_rate": 1.730826874642694e-05,
      "loss": 0.1766,
      "mean_token_accuracy": 0.8860064744949341,
      "num_tokens": 15654819.0,
      "step": 470
    }
  ],
  "logging_steps": 5,
  "max_steps": 474,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 8.406647877329224e+17,
  "train_batch_size": 28,
  "trial_name": null,
  "trial_params": null
}