{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9945344330716486,
  "eval_steps": 500,
  "global_step": 1257,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.023849746596442414,
      "grad_norm": 3.046875,
      "learning_rate": 2.368421052631579e-05,
      "loss": 1.8936,
      "mean_token_accuracy": 0.6058215032021205,
      "step": 10
    },
    {
      "epoch": 0.04769949319288483,
      "grad_norm": 1.7109375,
      "learning_rate": 5e-05,
      "loss": 1.5408,
      "mean_token_accuracy": 0.6348679741223653,
      "step": 20
    },
    {
      "epoch": 0.07154923978932724,
      "grad_norm": 1.6953125,
      "learning_rate": 7.631578947368422e-05,
      "loss": 1.4672,
      "mean_token_accuracy": 0.6460853961606821,
      "step": 30
    },
    {
      "epoch": 0.09539898638576966,
      "grad_norm": 1.546875,
      "learning_rate": 9.999983395259131e-05,
      "loss": 1.4522,
      "mean_token_accuracy": 0.6488149074216684,
      "step": 40
    },
    {
      "epoch": 0.11924873298221206,
      "grad_norm": 1.96875,
      "learning_rate": 9.997990959798673e-05,
      "loss": 1.431,
      "mean_token_accuracy": 0.6533399142324925,
      "step": 50
    },
    {
      "epoch": 0.14309847957865449,
      "grad_norm": 1.328125,
      "learning_rate": 9.99267909244468e-05,
      "loss": 1.4142,
      "mean_token_accuracy": 0.6570006487270196,
      "step": 60
    },
    {
      "epoch": 0.1669482261750969,
      "grad_norm": 1.6796875,
      "learning_rate": 9.984051321091072e-05,
      "loss": 1.3949,
      "mean_token_accuracy": 0.6594945053259532,
      "step": 70
    },
    {
      "epoch": 0.19079797277153931,
      "grad_norm": 1.4609375,
      "learning_rate": 9.972113375900154e-05,
      "loss": 1.3968,
      "mean_token_accuracy": 0.6598712412019571,
      "step": 80
    },
    {
      "epoch": 0.21464771936798172,
      "grad_norm": 1.7109375,
      "learning_rate": 9.956873185496916e-05,
      "loss": 1.387,
      "mean_token_accuracy": 0.6636203820506732,
      "step": 90
    },
    {
      "epoch": 0.23849746596442412,
      "grad_norm": 1.421875,
      "learning_rate": 9.9383408717032e-05,
      "loss": 1.3872,
      "mean_token_accuracy": 0.6624753788113594,
      "step": 100
    },
    {
      "epoch": 0.2623472125608665,
      "grad_norm": 1.171875,
      "learning_rate": 9.916528742815276e-05,
      "loss": 1.3453,
      "mean_token_accuracy": 0.6708123281598091,
      "step": 110
    },
    {
      "epoch": 0.28619695915730897,
      "grad_norm": 1.4921875,
      "learning_rate": 9.891451285429231e-05,
      "loss": 1.3512,
      "mean_token_accuracy": 0.6688777978221575,
      "step": 120
    },
    {
      "epoch": 0.3100467057537514,
      "grad_norm": 1.1171875,
      "learning_rate": 9.863125154819663e-05,
      "loss": 1.3764,
      "mean_token_accuracy": 0.6649044439196586,
      "step": 130
    },
    {
      "epoch": 0.3338964523501938,
      "grad_norm": 1.1875,
      "learning_rate": 9.831569163878015e-05,
      "loss": 1.3456,
      "mean_token_accuracy": 0.6716934656103452,
      "step": 140
    },
    {
      "epoch": 0.3577461989466362,
      "grad_norm": 1.2578125,
      "learning_rate": 9.796804270617937e-05,
      "loss": 1.3439,
      "mean_token_accuracy": 0.6696053599317868,
      "step": 150
    },
    {
      "epoch": 0.38159594554307863,
      "grad_norm": 1.0390625,
      "learning_rate": 9.758853564255954e-05,
      "loss": 1.3524,
      "mean_token_accuracy": 0.6700724132359028,
      "step": 160
    },
    {
      "epoch": 0.40544569213952103,
      "grad_norm": 1.1640625,
      "learning_rate": 9.71774224987669e-05,
      "loss": 1.3421,
      "mean_token_accuracy": 0.6707797152300675,
      "step": 170
    },
    {
      "epoch": 0.42929543873596343,
      "grad_norm": 1.0546875,
      "learning_rate": 9.673497631692821e-05,
      "loss": 1.3112,
      "mean_token_accuracy": 0.6768868687252204,
      "step": 180
    },
    {
      "epoch": 0.45314518533240583,
      "grad_norm": 1.0546875,
      "learning_rate": 9.626149094910905e-05,
      "loss": 1.3133,
      "mean_token_accuracy": 0.675601539760828,
      "step": 190
    },
    {
      "epoch": 0.47699493192884823,
      "grad_norm": 1.0703125,
      "learning_rate": 9.575728086215092e-05,
      "loss": 1.3186,
      "mean_token_accuracy": 0.6764184221625328,
      "step": 200
    },
    {
      "epoch": 0.5008446785252907,
      "grad_norm": 1.0625,
      "learning_rate": 9.522268092881716e-05,
      "loss": 1.34,
      "mean_token_accuracy": 0.6720153917868932,
      "step": 210
    },
    {
      "epoch": 0.524694425121733,
      "grad_norm": 1.2421875,
      "learning_rate": 9.465804620538601e-05,
      "loss": 1.3072,
      "mean_token_accuracy": 0.6793769620358944,
      "step": 220
    },
    {
      "epoch": 0.5485441717181755,
      "grad_norm": 1.109375,
      "learning_rate": 9.406375169583893e-05,
      "loss": 1.3028,
      "mean_token_accuracy": 0.6790211101373037,
      "step": 230
    },
    {
      "epoch": 0.5723939183146179,
      "grad_norm": 0.98828125,
      "learning_rate": 9.34401921028004e-05,
      "loss": 1.319,
      "mean_token_accuracy": 0.6775049904982249,
      "step": 240
    },
    {
      "epoch": 0.5962436649110603,
      "grad_norm": 0.95703125,
      "learning_rate": 9.278778156539502e-05,
      "loss": 1.2948,
      "mean_token_accuracy": 0.6797074491779009,
      "step": 250
    },
    {
      "epoch": 0.6200934115075027,
      "grad_norm": 0.91796875,
      "learning_rate": 9.210695338419553e-05,
      "loss": 1.2894,
      "mean_token_accuracy": 0.6807446469863255,
      "step": 260
    },
    {
      "epoch": 0.6439431581039451,
      "grad_norm": 0.9140625,
      "learning_rate": 9.139815973344493e-05,
      "loss": 1.3118,
      "mean_token_accuracy": 0.6776151955127716,
      "step": 270
    },
    {
      "epoch": 0.6677929047003875,
      "grad_norm": 1.03125,
      "learning_rate": 9.066187136074344e-05,
      "loss": 1.2922,
      "mean_token_accuracy": 0.6806117855012417,
      "step": 280
    },
    {
      "epoch": 0.69164265129683,
      "grad_norm": 0.98828125,
      "learning_rate": 8.989857727439994e-05,
      "loss": 1.2821,
      "mean_token_accuracy": 0.6829586572945118,
      "step": 290
    },
    {
      "epoch": 0.7154923978932723,
      "grad_norm": 1.0078125,
      "learning_rate": 8.91087844186556e-05,
      "loss": 1.2876,
      "mean_token_accuracy": 0.6828203638394673,
      "step": 300
    },
    {
      "epoch": 0.7393421444897148,
      "grad_norm": 0.94140625,
      "learning_rate": 8.829301733699511e-05,
      "loss": 1.2681,
      "mean_token_accuracy": 0.6843130300442378,
      "step": 310
    },
    {
      "epoch": 0.7631918910861573,
      "grad_norm": 1.015625,
      "learning_rate": 8.745181782376955e-05,
      "loss": 1.2828,
      "mean_token_accuracy": 0.6828058943152427,
      "step": 320
    },
    {
      "epoch": 0.7870416376825996,
      "grad_norm": 0.92578125,
      "learning_rate": 8.658574456436177e-05,
      "loss": 1.2724,
      "mean_token_accuracy": 0.6842800090710323,
      "step": 330
    },
    {
      "epoch": 0.8108913842790421,
      "grad_norm": 0.90234375,
      "learning_rate": 8.56953727641338e-05,
      "loss": 1.2904,
      "mean_token_accuracy": 0.6812497737507025,
      "step": 340
    },
    {
      "epoch": 0.8347411308754844,
      "grad_norm": 0.89453125,
      "learning_rate": 8.47812937664023e-05,
      "loss": 1.2815,
      "mean_token_accuracy": 0.685196939855814,
      "step": 350
    },
    {
      "epoch": 0.8585908774719269,
      "grad_norm": 0.95703125,
      "learning_rate": 8.384411465969597e-05,
      "loss": 1.263,
      "mean_token_accuracy": 0.6859518033762773,
      "step": 360
    },
    {
      "epoch": 0.8824406240683693,
      "grad_norm": 0.8828125,
      "learning_rate": 8.288445787455579e-05,
      "loss": 1.265,
      "mean_token_accuracy": 0.6860440624256928,
      "step": 370
    },
    {
      "epoch": 0.9062903706648117,
      "grad_norm": 0.90234375,
      "learning_rate": 8.190296077014563e-05,
      "loss": 1.277,
      "mean_token_accuracy": 0.6848570423821608,
      "step": 380
    },
    {
      "epoch": 0.9301401172612541,
      "grad_norm": 1.0078125,
      "learning_rate": 8.090027521094826e-05,
      "loss": 1.2564,
      "mean_token_accuracy": 0.6886172123253346,
      "step": 390
    },
    {
      "epoch": 0.9539898638576965,
      "grad_norm": 0.93359375,
      "learning_rate": 7.987706713382735e-05,
      "loss": 1.2191,
      "mean_token_accuracy": 0.694924907386303,
      "step": 400
    },
    {
      "epoch": 0.9778396104541389,
      "grad_norm": 0.921875,
      "learning_rate": 7.883401610574336e-05,
      "loss": 1.2559,
      "mean_token_accuracy": 0.6870772863427798,
      "step": 410
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.40625,
      "learning_rate": 7.777181487241691e-05,
      "loss": 1.2608,
      "mean_token_accuracy": 0.6875777412957674,
      "step": 420
    },
    {
      "epoch": 1.0238497465964425,
      "grad_norm": 0.95703125,
      "learning_rate": 7.669116889823955e-05,
      "loss": 1.0743,
      "mean_token_accuracy": 0.7243270235757033,
      "step": 430
    },
    {
      "epoch": 1.047699493192885,
      "grad_norm": 0.97265625,
      "learning_rate": 7.559279589773715e-05,
      "loss": 1.056,
      "mean_token_accuracy": 0.7269702054560184,
      "step": 440
    },
    {
      "epoch": 1.0715492397893271,
      "grad_norm": 1.0078125,
      "learning_rate": 7.447742535889768e-05,
      "loss": 1.0525,
      "mean_token_accuracy": 0.7275599343081315,
      "step": 450
    },
    {
      "epoch": 1.0953989863857696,
      "grad_norm": 0.98828125,
      "learning_rate": 7.334579805867918e-05,
      "loss": 1.0496,
      "mean_token_accuracy": 0.7283496469259262,
      "step": 460
    },
    {
      "epoch": 1.119248732982212,
      "grad_norm": 0.9765625,
      "learning_rate": 7.219866557102051e-05,
      "loss": 1.0534,
      "mean_token_accuracy": 0.7267930120229721,
      "step": 470
    },
    {
      "epoch": 1.1430984795786545,
      "grad_norm": 1.015625,
      "learning_rate": 7.103678976768102e-05,
      "loss": 1.0539,
      "mean_token_accuracy": 0.7286298322180907,
      "step": 480
    },
    {
      "epoch": 1.166948226175097,
      "grad_norm": 0.9921875,
      "learning_rate": 6.986094231224089e-05,
      "loss": 1.0589,
      "mean_token_accuracy": 0.7247064632674058,
      "step": 490
    },
    {
      "epoch": 1.1907979727715392,
      "grad_norm": 1.015625,
      "learning_rate": 6.867190414759835e-05,
      "loss": 1.0528,
      "mean_token_accuracy": 0.7296631435553232,
      "step": 500
    },
    {
      "epoch": 1.2146477193679817,
      "grad_norm": 0.91015625,
      "learning_rate": 6.747046497730395e-05,
      "loss": 1.067,
      "mean_token_accuracy": 0.7264716491103173,
      "step": 510
    },
    {
      "epoch": 1.2384974659644241,
      "grad_norm": 0.96484375,
      "learning_rate": 6.625742274107625e-05,
      "loss": 1.0602,
      "mean_token_accuracy": 0.7274638287723064,
      "step": 520
    },
    {
      "epoch": 1.2623472125608666,
      "grad_norm": 0.91015625,
      "learning_rate": 6.503358308484771e-05,
      "loss": 1.0547,
      "mean_token_accuracy": 0.7279857188463211,
      "step": 530
    },
    {
      "epoch": 1.286196959157309,
      "grad_norm": 1.0,
      "learning_rate": 6.379975882569224e-05,
      "loss": 1.0496,
      "mean_token_accuracy": 0.7278440314034621,
      "step": 540
    },
    {
      "epoch": 1.3100467057537513,
      "grad_norm": 0.9375,
      "learning_rate": 6.255676941199018e-05,
      "loss": 1.0517,
      "mean_token_accuracy": 0.7278454805413882,
      "step": 550
    },
    {
      "epoch": 1.3338964523501937,
      "grad_norm": 0.87109375,
      "learning_rate": 6.130544037918894e-05,
      "loss": 1.0465,
      "mean_token_accuracy": 0.7296144522726535,
      "step": 560
    },
    {
      "epoch": 1.3577461989466362,
      "grad_norm": 0.99609375,
      "learning_rate": 6.004660280152096e-05,
      "loss": 1.0441,
      "mean_token_accuracy": 0.7295484840869904,
      "step": 570
    },
    {
      "epoch": 1.3815959455430786,
      "grad_norm": 1.2890625,
      "learning_rate": 5.878109274004314e-05,
      "loss": 1.0399,
      "mean_token_accuracy": 0.7302657755712668,
      "step": 580
    },
    {
      "epoch": 1.405445692139521,
      "grad_norm": 0.93359375,
      "learning_rate": 5.7509750687364105e-05,
      "loss": 1.0245,
      "mean_token_accuracy": 0.7333451308310032,
      "step": 590
    },
    {
      "epoch": 1.4292954387359633,
      "grad_norm": 0.921875,
      "learning_rate": 5.62334210094283e-05,
      "loss": 1.0438,
      "mean_token_accuracy": 0.7301900732020538,
      "step": 600
    },
    {
      "epoch": 1.4531451853324058,
      "grad_norm": 1.0,
      "learning_rate": 5.495295138472769e-05,
      "loss": 1.0512,
      "mean_token_accuracy": 0.7292244640489419,
      "step": 610
    },
    {
      "epoch": 1.4769949319288482,
      "grad_norm": 0.94921875,
      "learning_rate": 5.366919224131331e-05,
      "loss": 1.0391,
      "mean_token_accuracy": 0.7299602496127288,
      "step": 620
    },
    {
      "epoch": 1.5008446785252907,
      "grad_norm": 0.9140625,
      "learning_rate": 5.238299619198066e-05,
      "loss": 1.048,
      "mean_token_accuracy": 0.7292950809001922,
      "step": 630
    },
    {
      "epoch": 1.5246944251217331,
      "grad_norm": 0.98828125,
      "learning_rate": 5.109521746800426e-05,
      "loss": 1.0443,
      "mean_token_accuracy": 0.729659582922856,
      "step": 640
    },
    {
      "epoch": 1.5485441717181754,
      "grad_norm": 0.90234375,
      "learning_rate": 4.980671135179708e-05,
      "loss": 1.0539,
      "mean_token_accuracy": 0.7284331674377124,
      "step": 650
    },
    {
      "epoch": 1.572393918314618,
      "grad_norm": 0.9453125,
      "learning_rate": 4.851833360887201e-05,
      "loss": 1.0498,
      "mean_token_accuracy": 0.7298707855244477,
      "step": 660
    },
    {
      "epoch": 1.5962436649110603,
      "grad_norm": 0.9296875,
      "learning_rate": 4.723093991948245e-05,
      "loss": 1.0226,
      "mean_token_accuracy": 0.7343422142167886,
      "step": 670
    },
    {
      "epoch": 1.6200934115075027,
      "grad_norm": 0.90625,
      "learning_rate": 4.5945385310319344e-05,
      "loss": 1.046,
      "mean_token_accuracy": 0.7300175433357556,
      "step": 680
    },
    {
      "epoch": 1.6439431581039452,
      "grad_norm": 0.91796875,
      "learning_rate": 4.466252358664269e-05,
      "loss": 1.0453,
      "mean_token_accuracy": 0.730091076095899,
      "step": 690
    },
    {
      "epoch": 1.6677929047003874,
      "grad_norm": 0.9453125,
      "learning_rate": 4.338320676522374e-05,
      "loss": 1.0366,
      "mean_token_accuracy": 0.732094819098711,
      "step": 700
    },
    {
      "epoch": 1.6916426512968301,
      "grad_norm": 0.87109375,
      "learning_rate": 4.210828450847551e-05,
      "loss": 1.0202,
      "mean_token_accuracy": 0.7349480445186297,
      "step": 710
    },
    {
      "epoch": 1.7154923978932723,
      "grad_norm": 0.890625,
      "learning_rate": 4.08386035601466e-05,
      "loss": 1.0287,
      "mean_token_accuracy": 0.7352750234305858,
      "step": 720
    },
    {
      "epoch": 1.7393421444897148,
      "grad_norm": 0.87890625,
      "learning_rate": 3.9575007182953633e-05,
      "loss": 1.0381,
      "mean_token_accuracy": 0.7321987241506577,
      "step": 730
    },
    {
      "epoch": 1.7631918910861573,
      "grad_norm": 1.015625,
      "learning_rate": 3.8318334598525625e-05,
      "loss": 1.0284,
      "mean_token_accuracy": 0.7348339579999447,
      "step": 740
    },
    {
      "epoch": 1.7870416376825995,
      "grad_norm": 0.95703125,
      "learning_rate": 3.706942043003215e-05,
      "loss": 1.0314,
      "mean_token_accuracy": 0.7322920034329097,
      "step": 750
    },
    {
      "epoch": 1.8108913842790422,
      "grad_norm": 0.859375,
      "learning_rate": 3.582909414786576e-05,
      "loss": 1.0316,
      "mean_token_accuracy": 0.7317293718457222,
      "step": 760
    },
    {
      "epoch": 1.8347411308754844,
      "grad_norm": 0.8984375,
      "learning_rate": 3.4598179518746375e-05,
      "loss": 1.0057,
      "mean_token_accuracy": 0.7380195428927739,
      "step": 770
    },
    {
      "epoch": 1.8585908774719269,
      "grad_norm": 0.8515625,
      "learning_rate": 3.3377494058614145e-05,
      "loss": 1.02,
      "mean_token_accuracy": 0.7366910368204117,
      "step": 780
    },
    {
      "epoch": 1.8824406240683693,
      "grad_norm": 0.8984375,
      "learning_rate": 3.216784848967341e-05,
      "loss": 1.0053,
      "mean_token_accuracy": 0.7390421097477277,
      "step": 790
    },
    {
      "epoch": 1.9062903706648116,
      "grad_norm": 0.8671875,
      "learning_rate": 3.097004620194902e-05,
      "loss": 1.0249,
      "mean_token_accuracy": 0.7343820112446944,
      "step": 800
    },
    {
      "epoch": 1.9301401172612542,
      "grad_norm": 0.8828125,
      "learning_rate": 2.9784882719712248e-05,
      "loss": 1.023,
      "mean_token_accuracy": 0.7356185607612133,
      "step": 810
    },
    {
      "epoch": 1.9539898638576965,
      "grad_norm": 0.9453125,
      "learning_rate": 2.861314517313071e-05,
      "loss": 1.0223,
      "mean_token_accuracy": 0.736139855782191,
      "step": 820
    },
    {
      "epoch": 1.977839610454139,
      "grad_norm": 0.890625,
      "learning_rate": 2.7455611775493363e-05,
      "loss": 1.0058,
      "mean_token_accuracy": 0.7381812597314517,
      "step": 830
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6796875,
      "learning_rate": 2.6313051306357583e-05,
      "loss": 1.0083,
      "mean_token_accuracy": 0.7398098124516919,
      "step": 840
    },
    {
      "epoch": 2.0238497465964422,
      "grad_norm": 1.2578125,
      "learning_rate": 2.5186222600961785e-05,
      "loss": 0.8511,
      "mean_token_accuracy": 0.7758485838770867,
      "step": 850
    },
    {
      "epoch": 2.047699493192885,
      "grad_norm": 0.9453125,
      "learning_rate": 2.407587404624253e-05,
      "loss": 0.8605,
      "mean_token_accuracy": 0.7730280997852484,
      "step": 860
    },
    {
      "epoch": 2.071549239789327,
      "grad_norm": 1.0703125,
      "learning_rate": 2.2982743083790858e-05,
      "loss": 0.843,
      "mean_token_accuracy": 0.7769588189820449,
      "step": 870
    },
    {
      "epoch": 2.09539898638577,
      "grad_norm": 1.0390625,
      "learning_rate": 2.1907555720078115e-05,
      "loss": 0.8554,
      "mean_token_accuracy": 0.7744472898542881,
      "step": 880
    },
    {
      "epoch": 2.119248732982212,
      "grad_norm": 0.9296875,
      "learning_rate": 2.0851026044276406e-05,
      "loss": 0.8534,
      "mean_token_accuracy": 0.7748738661408424,
      "step": 890
    },
    {
      "epoch": 2.1430984795786543,
      "grad_norm": 0.94140625,
      "learning_rate": 1.9813855753993914e-05,
      "loss": 0.8549,
      "mean_token_accuracy": 0.7733134960134824,
      "step": 900
    },
    {
      "epoch": 2.166948226175097,
      "grad_norm": 0.9765625,
      "learning_rate": 1.879673368924025e-05,
      "loss": 0.8385,
      "mean_token_accuracy": 0.7773586166401704,
      "step": 910
    },
    {
      "epoch": 2.190797972771539,
      "grad_norm": 0.9375,
      "learning_rate": 1.7800335374930998e-05,
      "loss": 0.8688,
      "mean_token_accuracy": 0.7720932170748711,
      "step": 920
    },
    {
      "epoch": 2.214647719367982,
      "grad_norm": 0.9765625,
      "learning_rate": 1.6825322572235767e-05,
      "loss": 0.8432,
      "mean_token_accuracy": 0.7758201735715071,
      "step": 930
    },
    {
      "epoch": 2.238497465964424,
      "grad_norm": 0.93359375,
      "learning_rate": 1.5872342839067306e-05,
      "loss": 0.8467,
      "mean_token_accuracy": 0.7756141935785611,
      "step": 940
    },
    {
      "epoch": 2.2623472125608663,
      "grad_norm": 0.9609375,
      "learning_rate": 1.4942029100003808e-05,
      "loss": 0.8615,
      "mean_token_accuracy": 0.7729161617656549,
      "step": 950
    },
    {
      "epoch": 2.286196959157309,
      "grad_norm": 0.9296875,
      "learning_rate": 1.4034999225930107e-05,
      "loss": 0.8604,
      "mean_token_accuracy": 0.7723979363838832,
      "step": 960
    },
    {
      "epoch": 2.3100467057537513,
      "grad_norm": 0.97265625,
      "learning_rate": 1.3151855623676517e-05,
      "loss": 0.8512,
      "mean_token_accuracy": 0.7750633083283901,
      "step": 970
    },
    {
      "epoch": 2.333896452350194,
      "grad_norm": 0.921875,
      "learning_rate": 1.2293184835928639e-05,
      "loss": 0.8555,
      "mean_token_accuracy": 0.7741216157873472,
      "step": 980
    },
    {
      "epoch": 2.357746198946636,
      "grad_norm": 0.96875,
      "learning_rate": 1.1459557151672933e-05,
      "loss": 0.8452,
      "mean_token_accuracy": 0.7758361441393693,
      "step": 990
    },
    {
      "epoch": 2.3815959455430784,
      "grad_norm": 0.92578125,
      "learning_rate": 1.065152622743753e-05,
      "loss": 0.8344,
      "mean_token_accuracy": 0.7779088502128919,
      "step": 1000
    },
    {
      "epoch": 2.405445692139521,
      "grad_norm": 0.95703125,
      "learning_rate": 9.869628719579448e-06,
      "loss": 0.8474,
      "mean_token_accuracy": 0.7763092644512654,
      "step": 1010
    },
    {
      "epoch": 2.4292954387359633,
      "grad_norm": 0.921875,
      "learning_rate": 9.114383927862502e-06,
      "loss": 0.8307,
      "mean_token_accuracy": 0.7789155239860217,
      "step": 1020
    },
    {
      "epoch": 2.453145185332406,
      "grad_norm": 0.9453125,
      "learning_rate": 8.386293450562755e-06,
      "loss": 0.8316,
      "mean_token_accuracy": 0.7797198290626208,
      "step": 1030
    },
    {
      "epoch": 2.4769949319288482,
      "grad_norm": 0.9296875,
      "learning_rate": 7.685840851330296e-06,
      "loss": 0.8276,
      "mean_token_accuracy": 0.7800254955887794,
      "step": 1040
    },
    {
      "epoch": 2.500844678525291,
      "grad_norm": 0.94921875,
      "learning_rate": 7.013491338028938e-06,
      "loss": 0.858,
      "mean_token_accuracy": 0.7726903068522613,
      "step": 1050
    },
    {
      "epoch": 2.524694425121733,
      "grad_norm": 0.921875,
      "learning_rate": 6.3696914537669184e-06,
      "loss": 0.8433,
      "mean_token_accuracy": 0.7763248644769192,
      "step": 1060
    },
    {
      "epoch": 2.5485441717181754,
      "grad_norm": 0.953125,
      "learning_rate": 5.7548687803237915e-06,
      "loss": 0.8508,
      "mean_token_accuracy": 0.7756962545216084,
      "step": 1070
    },
    {
      "epoch": 2.572393918314618,
      "grad_norm": 0.9453125,
      "learning_rate": 5.16943165417067e-06,
      "loss": 0.8344,
      "mean_token_accuracy": 0.7778078898787498,
      "step": 1080
    },
    {
      "epoch": 2.5962436649110603,
      "grad_norm": 0.91796875,
      "learning_rate": 4.613768895272181e-06,
      "loss": 0.8511,
      "mean_token_accuracy": 0.7750198910633723,
      "step": 1090
    },
    {
      "epoch": 2.6200934115075025,
      "grad_norm": 0.92578125,
      "learning_rate": 4.088249548850442e-06,
      "loss": 0.8429,
      "mean_token_accuracy": 0.7772863209247589,
      "step": 1100
    },
    {
      "epoch": 2.643943158103945,
      "grad_norm": 0.98046875,
      "learning_rate": 3.5932226402824954e-06,
      "loss": 0.8517,
      "mean_token_accuracy": 0.7759914328654607,
      "step": 1110
    },
    {
      "epoch": 2.6677929047003874,
      "grad_norm": 0.95703125,
      "learning_rate": 3.1290169432939553e-06,
      "loss": 0.8544,
      "mean_token_accuracy": 0.7765047900378704,
      "step": 1120
    },
    {
      "epoch": 2.69164265129683,
      "grad_norm": 0.94140625,
      "learning_rate": 2.6959407616028997e-06,
      "loss": 0.8327,
      "mean_token_accuracy": 0.778083398193121,
      "step": 1130
    },
    {
      "epoch": 2.7154923978932723,
      "grad_norm": 0.953125,
      "learning_rate": 2.294281724158892e-06,
      "loss": 0.8517,
      "mean_token_accuracy": 0.7756194410224756,
      "step": 1140
    },
    {
      "epoch": 2.739342144489715,
      "grad_norm": 1.5390625,
      "learning_rate": 1.9243065941133787e-06,
      "loss": 0.8394,
      "mean_token_accuracy": 0.7775321317215761,
      "step": 1150
    },
    {
      "epoch": 2.7631918910861573,
      "grad_norm": 0.91796875,
      "learning_rate": 1.5862610916479893e-06,
      "loss": 0.8432,
      "mean_token_accuracy": 0.7762146083017191,
      "step": 1160
    },
    {
      "epoch": 2.7870416376825995,
      "grad_norm": 0.91796875,
      "learning_rate": 1.280369730778741e-06,
      "loss": 0.8434,
      "mean_token_accuracy": 0.7769906096160412,
      "step": 1170
    },
    {
      "epoch": 2.810891384279042,
      "grad_norm": 1.1796875,
      "learning_rate": 1.0068356702442827e-06,
      "loss": 0.8558,
      "mean_token_accuracy": 0.7744560035566489,
      "step": 1180
    },
    {
      "epoch": 2.8347411308754844,
      "grad_norm": 0.93359375,
      "learning_rate": 7.658405785773926e-07,
      "loss": 0.8373,
      "mean_token_accuracy": 0.7771262871722381,
      "step": 1190
    },
    {
      "epoch": 2.8585908774719266,
      "grad_norm": 0.953125,
      "learning_rate": 5.575445134492374e-07,
      "loss": 0.8336,
      "mean_token_accuracy": 0.7793849507967631,
      "step": 1200
    },
    {
      "epoch": 2.8824406240683693,
      "grad_norm": 0.94921875,
      "learning_rate": 3.8208581536650635e-07,
      "loss": 0.8291,
      "mean_token_accuracy": 0.7792126409709453,
      "step": 1210
    },
    {
      "epoch": 2.9062903706648116,
      "grad_norm": 0.90234375,
      "learning_rate": 2.395810157921674e-07,
      "loss": 0.832,
      "mean_token_accuracy": 0.7793585898975531,
      "step": 1220
    },
    {
      "epoch": 2.9301401172612542,
      "grad_norm": 0.94140625,
      "learning_rate": 1.3012475975068828e-07,
      "loss": 0.8489,
      "mean_token_accuracy": 0.7747443556785584,
      "step": 1230
    },
    {
      "epoch": 2.9539898638576965,
      "grad_norm": 0.94921875,
      "learning_rate": 5.378974296924577e-08,
      "loss": 0.829,
      "mean_token_accuracy": 0.7786815270781517,
      "step": 1240
    },
    {
      "epoch": 2.977839610454139,
      "grad_norm": 1.0234375,
      "learning_rate": 1.0626663596624564e-08,
      "loss": 0.8445,
      "mean_token_accuracy": 0.7772307982047398,
      "step": 1250
    }
  ],
  "logging_steps": 10,
  "max_steps": 1257,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.602088301310771e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}