{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 215,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004662004662004662,
      "grad_norm": 106.21962531797728,
      "learning_rate": 0.0,
      "loss": 2.739,
      "mean_token_accuracy": 0.5701408684253693,
      "num_tokens": 9481.0,
      "step": 1
    },
    {
      "epoch": 0.009324009324009324,
      "grad_norm": 113.36914382118674,
      "learning_rate": 4.5454545454545457e-07,
      "loss": 2.82,
      "mean_token_accuracy": 0.56001877784729,
      "num_tokens": 18486.0,
      "step": 2
    },
    {
      "epoch": 0.013986013986013986,
      "grad_norm": 106.50317238622108,
      "learning_rate": 9.090909090909091e-07,
      "loss": 2.6926,
      "mean_token_accuracy": 0.5815096497535706,
      "num_tokens": 28054.0,
      "step": 3
    },
    {
      "epoch": 0.018648018648018648,
      "grad_norm": 90.02623522903296,
      "learning_rate": 1.3636363636363636e-06,
      "loss": 2.5138,
      "mean_token_accuracy": 0.5946728587150574,
      "num_tokens": 38100.0,
      "step": 4
    },
    {
      "epoch": 0.023310023310023312,
      "grad_norm": 77.42719349366115,
      "learning_rate": 1.8181818181818183e-06,
      "loss": 2.581,
      "mean_token_accuracy": 0.5861297249794006,
      "num_tokens": 46674.0,
      "step": 5
    },
    {
      "epoch": 0.027972027972027972,
      "grad_norm": 51.84205661670952,
      "learning_rate": 2.2727272727272728e-06,
      "loss": 2.0848,
      "mean_token_accuracy": 0.6379857063293457,
      "num_tokens": 55892.0,
      "step": 6
    },
    {
      "epoch": 0.03263403263403263,
      "grad_norm": 47.88064411349626,
      "learning_rate": 2.7272727272727272e-06,
      "loss": 1.9382,
      "mean_token_accuracy": 0.6385776698589325,
      "num_tokens": 64557.0,
      "step": 7
    },
    {
      "epoch": 0.037296037296037296,
      "grad_norm": 31.663436012051196,
      "learning_rate": 3.181818181818182e-06,
      "loss": 1.4512,
      "mean_token_accuracy": 0.7073646187782288,
      "num_tokens": 74534.0,
      "step": 8
    },
    {
      "epoch": 0.04195804195804196,
      "grad_norm": 28.379675867254864,
      "learning_rate": 3.6363636363636366e-06,
      "loss": 1.273,
      "mean_token_accuracy": 0.7217381000518799,
      "num_tokens": 84839.0,
      "step": 9
    },
    {
      "epoch": 0.046620046620046623,
      "grad_norm": 22.05633895747606,
      "learning_rate": 4.0909090909090915e-06,
      "loss": 0.9671,
      "mean_token_accuracy": 0.7773665189743042,
      "num_tokens": 94486.0,
      "step": 10
    },
    {
      "epoch": 0.05128205128205128,
      "grad_norm": 12.286554458797058,
      "learning_rate": 4.5454545454545455e-06,
      "loss": 0.7773,
      "mean_token_accuracy": 0.8185156285762787,
      "num_tokens": 103901.0,
      "step": 11
    },
    {
      "epoch": 0.055944055944055944,
      "grad_norm": 8.71894709105133,
      "learning_rate": 5e-06,
      "loss": 0.7136,
      "mean_token_accuracy": 0.8239960074424744,
      "num_tokens": 113421.0,
      "step": 12
    },
    {
      "epoch": 0.06060606060606061,
      "grad_norm": 7.079633705012988,
      "learning_rate": 5.4545454545454545e-06,
      "loss": 0.6871,
      "mean_token_accuracy": 0.8277380168437958,
      "num_tokens": 122888.0,
      "step": 13
    },
    {
      "epoch": 0.06526806526806526,
      "grad_norm": 5.438376413266983,
      "learning_rate": 5.90909090909091e-06,
      "loss": 0.6091,
      "mean_token_accuracy": 0.8382397592067719,
      "num_tokens": 132046.0,
      "step": 14
    },
    {
      "epoch": 0.06993006993006994,
      "grad_norm": 3.3812206364126145,
      "learning_rate": 6.363636363636364e-06,
      "loss": 0.5661,
      "mean_token_accuracy": 0.854902058839798,
      "num_tokens": 141586.0,
      "step": 15
    },
    {
      "epoch": 0.07459207459207459,
      "grad_norm": 3.456199391812718,
      "learning_rate": 6.818181818181818e-06,
      "loss": 0.5344,
      "mean_token_accuracy": 0.8581095337867737,
      "num_tokens": 151366.0,
      "step": 16
    },
    {
      "epoch": 0.07925407925407925,
      "grad_norm": 4.262280791274996,
      "learning_rate": 7.272727272727273e-06,
      "loss": 0.5444,
      "mean_token_accuracy": 0.8576282858848572,
      "num_tokens": 160361.0,
      "step": 17
    },
    {
      "epoch": 0.08391608391608392,
      "grad_norm": 3.50951187171187,
      "learning_rate": 7.727272727272727e-06,
      "loss": 0.5512,
      "mean_token_accuracy": 0.8535145223140717,
      "num_tokens": 170716.0,
      "step": 18
    },
    {
      "epoch": 0.08857808857808858,
      "grad_norm": 3.1077568547014907,
      "learning_rate": 8.181818181818183e-06,
      "loss": 0.564,
      "mean_token_accuracy": 0.8447403013706207,
      "num_tokens": 179378.0,
      "step": 19
    },
    {
      "epoch": 0.09324009324009325,
      "grad_norm": 2.9392782111385554,
      "learning_rate": 8.636363636363637e-06,
      "loss": 0.4988,
      "mean_token_accuracy": 0.8656269907951355,
      "num_tokens": 189155.0,
      "step": 20
    },
    {
      "epoch": 0.0979020979020979,
      "grad_norm": 3.014369371976925,
      "learning_rate": 9.090909090909091e-06,
      "loss": 0.5534,
      "mean_token_accuracy": 0.8481437265872955,
      "num_tokens": 198128.0,
      "step": 21
    },
    {
      "epoch": 0.10256410256410256,
      "grad_norm": 3.2412985537099632,
      "learning_rate": 9.545454545454547e-06,
      "loss": 0.536,
      "mean_token_accuracy": 0.855196475982666,
      "num_tokens": 207526.0,
      "step": 22
    },
    {
      "epoch": 0.10722610722610723,
      "grad_norm": 2.8906738942062002,
      "learning_rate": 1e-05,
      "loss": 0.5493,
      "mean_token_accuracy": 0.8555976152420044,
      "num_tokens": 217437.0,
      "step": 23
    },
    {
      "epoch": 0.11188811188811189,
      "grad_norm": 2.8480052706011234,
      "learning_rate": 9.999403846557509e-06,
      "loss": 0.5421,
      "mean_token_accuracy": 0.8500343859195709,
      "num_tokens": 226725.0,
      "step": 24
    },
    {
      "epoch": 0.11655011655011654,
      "grad_norm": 2.7151330990573572,
      "learning_rate": 9.99761554418511e-06,
      "loss": 0.4959,
      "mean_token_accuracy": 0.8620002269744873,
      "num_tokens": 235721.0,
      "step": 25
    },
    {
      "epoch": 0.12121212121212122,
      "grad_norm": 2.9193258742656867,
      "learning_rate": 9.99463556670619e-06,
      "loss": 0.5549,
      "mean_token_accuracy": 0.8563002645969391,
      "num_tokens": 244662.0,
      "step": 26
    },
    {
      "epoch": 0.1258741258741259,
      "grad_norm": 2.9856041530329422,
      "learning_rate": 9.990464703686895e-06,
      "loss": 0.5182,
      "mean_token_accuracy": 0.858866274356842,
      "num_tokens": 254047.0,
      "step": 27
    },
    {
      "epoch": 0.13053613053613053,
      "grad_norm": 2.931527022702967,
      "learning_rate": 9.985104060226937e-06,
      "loss": 0.5415,
      "mean_token_accuracy": 0.8544901311397552,
      "num_tokens": 262770.0,
      "step": 28
    },
    {
      "epoch": 0.1351981351981352,
      "grad_norm": 2.913225747288928,
      "learning_rate": 9.978555056666784e-06,
      "loss": 0.519,
      "mean_token_accuracy": 0.8577743768692017,
      "num_tokens": 272936.0,
      "step": 29
    },
    {
      "epoch": 0.13986013986013987,
      "grad_norm": 3.1388325596538524,
      "learning_rate": 9.97081942821133e-06,
      "loss": 0.5488,
      "mean_token_accuracy": 0.8508030474185944,
      "num_tokens": 282677.0,
      "step": 30
    },
    {
      "epoch": 0.1445221445221445,
      "grad_norm": 2.8336187119045966,
      "learning_rate": 9.961899224470146e-06,
      "loss": 0.5551,
      "mean_token_accuracy": 0.8455260097980499,
      "num_tokens": 291851.0,
      "step": 31
    },
    {
      "epoch": 0.14918414918414918,
      "grad_norm": 2.8464196245759834,
      "learning_rate": 9.95179680891442e-06,
      "loss": 0.5215,
      "mean_token_accuracy": 0.8570860624313354,
      "num_tokens": 300775.0,
      "step": 32
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 3.1695581640338353,
      "learning_rate": 9.940514858250736e-06,
      "loss": 0.5782,
      "mean_token_accuracy": 0.843131422996521,
      "num_tokens": 309003.0,
      "step": 33
    },
    {
      "epoch": 0.1585081585081585,
      "grad_norm": 2.648702859970465,
      "learning_rate": 9.928056361711854e-06,
      "loss": 0.5384,
      "mean_token_accuracy": 0.8559990227222443,
      "num_tokens": 318722.0,
      "step": 34
    },
    {
      "epoch": 0.16317016317016317,
      "grad_norm": 2.959958214770934,
      "learning_rate": 9.914424620264714e-06,
      "loss": 0.5722,
      "mean_token_accuracy": 0.842935174703598,
      "num_tokens": 328363.0,
      "step": 35
    },
    {
      "epoch": 0.16783216783216784,
      "grad_norm": 2.8111816604022652,
      "learning_rate": 9.899623245735798e-06,
      "loss": 0.5593,
      "mean_token_accuracy": 0.848853588104248,
      "num_tokens": 337298.0,
      "step": 36
    },
    {
      "epoch": 0.17249417249417248,
      "grad_norm": 2.689318326683542,
      "learning_rate": 9.883656159854166e-06,
      "loss": 0.514,
      "mean_token_accuracy": 0.8601629137992859,
      "num_tokens": 346068.0,
      "step": 37
    },
    {
      "epoch": 0.17715617715617715,
      "grad_norm": 2.6642622973952856,
      "learning_rate": 9.866527593212355e-06,
      "loss": 0.5347,
      "mean_token_accuracy": 0.85555499792099,
      "num_tokens": 354566.0,
      "step": 38
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 2.4952746525538556,
      "learning_rate": 9.848242084145462e-06,
      "loss": 0.4753,
      "mean_token_accuracy": 0.8724182546138763,
      "num_tokens": 363662.0,
      "step": 39
    },
    {
      "epoch": 0.1864801864801865,
      "grad_norm": 2.844156691505105,
      "learning_rate": 9.82880447752868e-06,
      "loss": 0.5545,
      "mean_token_accuracy": 0.8522640764713287,
      "num_tokens": 372778.0,
      "step": 40
    },
    {
      "epoch": 0.19114219114219114,
      "grad_norm": 2.830813192153553,
      "learning_rate": 9.808219923493606e-06,
      "loss": 0.5447,
      "mean_token_accuracy": 0.8559394776821136,
      "num_tokens": 381990.0,
      "step": 41
    },
    {
      "epoch": 0.1958041958041958,
      "grad_norm": 2.9359560870526633,
      "learning_rate": 9.786493876063685e-06,
      "loss": 0.5475,
      "mean_token_accuracy": 0.8542667627334595,
      "num_tokens": 390764.0,
      "step": 42
    },
    {
      "epoch": 0.20046620046620048,
      "grad_norm": 2.6640953874089237,
      "learning_rate": 9.763632091709125e-06,
      "loss": 0.5008,
      "mean_token_accuracy": 0.8638215363025665,
      "num_tokens": 399544.0,
      "step": 43
    },
    {
      "epoch": 0.20512820512820512,
      "grad_norm": 2.734444501240376,
      "learning_rate": 9.739640627821678e-06,
      "loss": 0.5612,
      "mean_token_accuracy": 0.8510912954807281,
      "num_tokens": 408895.0,
      "step": 44
    },
    {
      "epoch": 0.2097902097902098,
      "grad_norm": 2.656700776297115,
      "learning_rate": 9.714525841109697e-06,
      "loss": 0.5091,
      "mean_token_accuracy": 0.8665460348129272,
      "num_tokens": 417528.0,
      "step": 45
    },
    {
      "epoch": 0.21445221445221446,
      "grad_norm": 3.373802699282139,
      "learning_rate": 9.68829438591387e-06,
      "loss": 0.5306,
      "mean_token_accuracy": 0.8545205891132355,
      "num_tokens": 426632.0,
      "step": 46
    },
    {
      "epoch": 0.2191142191142191,
      "grad_norm": 2.8176592854523705,
      "learning_rate": 9.660953212444116e-06,
      "loss": 0.5676,
      "mean_token_accuracy": 0.8484572172164917,
      "num_tokens": 435698.0,
      "step": 47
    },
    {
      "epoch": 0.22377622377622378,
      "grad_norm": 2.5757476610659764,
      "learning_rate": 9.632509564938073e-06,
      "loss": 0.5128,
      "mean_token_accuracy": 0.8621232509613037,
      "num_tokens": 444865.0,
      "step": 48
    },
    {
      "epoch": 0.22843822843822845,
      "grad_norm": 2.636964562018401,
      "learning_rate": 9.60297097974169e-06,
      "loss": 0.5074,
      "mean_token_accuracy": 0.8599043786525726,
      "num_tokens": 454293.0,
      "step": 49
    },
    {
      "epoch": 0.2331002331002331,
      "grad_norm": 2.4595885234667842,
      "learning_rate": 9.572345283312407e-06,
      "loss": 0.5375,
      "mean_token_accuracy": 0.8563795983791351,
      "num_tokens": 463669.0,
      "step": 50
    },
    {
      "epoch": 0.23776223776223776,
      "grad_norm": 2.9425482274953807,
      "learning_rate": 9.540640590145496e-06,
      "loss": 0.5524,
      "mean_token_accuracy": 0.8539282977581024,
      "num_tokens": 472913.0,
      "step": 51
    },
    {
      "epoch": 0.24242424242424243,
      "grad_norm": 2.6095415458131836,
      "learning_rate": 9.507865300624057e-06,
      "loss": 0.5173,
      "mean_token_accuracy": 0.8600243926048279,
      "num_tokens": 482966.0,
      "step": 52
    },
    {
      "epoch": 0.24708624708624707,
      "grad_norm": 2.681507115816294,
      "learning_rate": 9.474028098793277e-06,
      "loss": 0.5332,
      "mean_token_accuracy": 0.8583089411258698,
      "num_tokens": 491361.0,
      "step": 53
    },
    {
      "epoch": 0.2517482517482518,
      "grad_norm": 2.385759042200182,
      "learning_rate": 9.439137950059539e-06,
      "loss": 0.5163,
      "mean_token_accuracy": 0.8575912415981293,
      "num_tokens": 500582.0,
      "step": 54
    },
    {
      "epoch": 0.2564102564102564,
      "grad_norm": 2.5465629384343944,
      "learning_rate": 9.403204098814965e-06,
      "loss": 0.5137,
      "mean_token_accuracy": 0.8564550578594208,
      "num_tokens": 510444.0,
      "step": 55
    },
    {
      "epoch": 0.26107226107226106,
      "grad_norm": 2.638187444461748,
      "learning_rate": 9.366236065988053e-06,
      "loss": 0.5207,
      "mean_token_accuracy": 0.8607774972915649,
      "num_tokens": 519595.0,
      "step": 56
    },
    {
      "epoch": 0.26573426573426573,
      "grad_norm": 2.625983528806919,
      "learning_rate": 9.32824364652104e-06,
      "loss": 0.505,
      "mean_token_accuracy": 0.861905574798584,
      "num_tokens": 529256.0,
      "step": 57
    },
    {
      "epoch": 0.2703962703962704,
      "grad_norm": 2.352386778290299,
      "learning_rate": 9.289236906774663e-06,
      "loss": 0.5261,
      "mean_token_accuracy": 0.8581192195415497,
      "num_tokens": 538883.0,
      "step": 58
    },
    {
      "epoch": 0.27505827505827507,
      "grad_norm": 2.513629377813567,
      "learning_rate": 9.249226181861e-06,
      "loss": 0.5102,
      "mean_token_accuracy": 0.8652426302433014,
      "num_tokens": 547820.0,
      "step": 59
    },
    {
      "epoch": 0.27972027972027974,
      "grad_norm": 2.5518094476836115,
      "learning_rate": 9.208222072905113e-06,
      "loss": 0.5169,
      "mean_token_accuracy": 0.8610005676746368,
      "num_tokens": 556572.0,
      "step": 60
    },
    {
      "epoch": 0.28438228438228436,
      "grad_norm": 2.6584992361477977,
      "learning_rate": 9.166235444236209e-06,
      "loss": 0.5414,
      "mean_token_accuracy": 0.8505589067935944,
      "num_tokens": 565988.0,
      "step": 61
    },
    {
      "epoch": 0.289044289044289,
      "grad_norm": 2.6790070596686415,
      "learning_rate": 9.123277420509053e-06,
      "loss": 0.5421,
      "mean_token_accuracy": 0.853904128074646,
      "num_tokens": 575388.0,
      "step": 62
    },
    {
      "epoch": 0.2937062937062937,
      "grad_norm": 2.645384290880974,
      "learning_rate": 9.079359383756411e-06,
      "loss": 0.5758,
      "mean_token_accuracy": 0.8486554026603699,
      "num_tokens": 584991.0,
      "step": 63
    },
    {
      "epoch": 0.29836829836829837,
      "grad_norm": 2.5250430356867466,
      "learning_rate": 9.034492970373305e-06,
      "loss": 0.5289,
      "mean_token_accuracy": 0.8559047877788544,
      "num_tokens": 594666.0,
      "step": 64
    },
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 2.4921775989926704,
      "learning_rate": 8.988690068033864e-06,
      "loss": 0.5389,
      "mean_token_accuracy": 0.8566691875457764,
      "num_tokens": 603192.0,
      "step": 65
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.5475631559795113,
      "learning_rate": 8.941962812541604e-06,
      "loss": 0.5219,
      "mean_token_accuracy": 0.8545815050601959,
      "num_tokens": 612501.0,
      "step": 66
    },
    {
      "epoch": 0.3123543123543124,
      "grad_norm": 2.7404611663741187,
      "learning_rate": 8.894323584613951e-06,
      "loss": 0.5343,
      "mean_token_accuracy": 0.8563965559005737,
      "num_tokens": 621564.0,
      "step": 67
    },
    {
      "epoch": 0.317016317016317,
      "grad_norm": 2.5161087036718213,
      "learning_rate": 8.845785006601898e-06,
      "loss": 0.4811,
      "mean_token_accuracy": 0.8693608343601227,
      "num_tokens": 630726.0,
      "step": 68
    },
    {
      "epoch": 0.32167832167832167,
      "grad_norm": 2.373054960197776,
      "learning_rate": 8.796359939145614e-06,
      "loss": 0.4981,
      "mean_token_accuracy": 0.864752858877182,
      "num_tokens": 639950.0,
      "step": 69
    },
    {
      "epoch": 0.32634032634032634,
      "grad_norm": 2.5926555831182303,
      "learning_rate": 8.74606147776692e-06,
      "loss": 0.5095,
      "mean_token_accuracy": 0.8598109483718872,
      "num_tokens": 649402.0,
      "step": 70
    },
    {
      "epoch": 0.331002331002331,
      "grad_norm": 2.5261642541122638,
      "learning_rate": 8.694902949399555e-06,
      "loss": 0.4848,
      "mean_token_accuracy": 0.8674550652503967,
      "num_tokens": 658474.0,
      "step": 71
    },
    {
      "epoch": 0.3356643356643357,
      "grad_norm": 2.552689438449325,
      "learning_rate": 8.642897908858096e-06,
      "loss": 0.5244,
      "mean_token_accuracy": 0.8566029667854309,
      "num_tokens": 667908.0,
      "step": 72
    },
    {
      "epoch": 0.34032634032634035,
      "grad_norm": 2.366696196458211,
      "learning_rate": 8.590060135246516e-06,
      "loss": 0.4982,
      "mean_token_accuracy": 0.8606800734996796,
      "num_tokens": 677532.0,
      "step": 73
    },
    {
      "epoch": 0.34498834498834496,
      "grad_norm": 2.4483720564534113,
      "learning_rate": 8.53640362830732e-06,
      "loss": 0.4972,
      "mean_token_accuracy": 0.8640582263469696,
      "num_tokens": 686796.0,
      "step": 74
    },
    {
      "epoch": 0.34965034965034963,
      "grad_norm": 2.4220030766893217,
      "learning_rate": 8.481942604712209e-06,
      "loss": 0.5144,
      "mean_token_accuracy": 0.8601466417312622,
      "num_tokens": 695776.0,
      "step": 75
    },
    {
      "epoch": 0.3543123543123543,
      "grad_norm": 2.4508536566710126,
      "learning_rate": 8.426691494295269e-06,
      "loss": 0.5292,
      "mean_token_accuracy": 0.8619424700737,
      "num_tokens": 704530.0,
      "step": 76
    },
    {
      "epoch": 0.358974358974359,
      "grad_norm": 2.3974226437864568,
      "learning_rate": 8.370664936229688e-06,
      "loss": 0.5145,
      "mean_token_accuracy": 0.8591651022434235,
      "num_tokens": 714123.0,
      "step": 77
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 2.6397896175282898,
      "learning_rate": 8.313877775149009e-06,
      "loss": 0.5124,
      "mean_token_accuracy": 0.8615582883358002,
      "num_tokens": 723719.0,
      "step": 78
    },
    {
      "epoch": 0.3682983682983683,
      "grad_norm": 2.6678029454235683,
      "learning_rate": 8.256345057213925e-06,
      "loss": 0.5372,
      "mean_token_accuracy": 0.8539088070392609,
      "num_tokens": 733447.0,
      "step": 79
    },
    {
      "epoch": 0.372960372960373,
      "grad_norm": 2.481956729573453,
      "learning_rate": 8.198082026125707e-06,
      "loss": 0.522,
      "mean_token_accuracy": 0.8558786809444427,
      "num_tokens": 742567.0,
      "step": 80
    },
    {
      "epoch": 0.3776223776223776,
      "grad_norm": 2.4651078936200856,
      "learning_rate": 8.139104119087265e-06,
      "loss": 0.5201,
      "mean_token_accuracy": 0.8631138801574707,
      "num_tokens": 751488.0,
      "step": 81
    },
    {
      "epoch": 0.3822843822843823,
      "grad_norm": 2.342231108488541,
      "learning_rate": 8.07942696271296e-06,
      "loss": 0.5112,
      "mean_token_accuracy": 0.8625693023204803,
      "num_tokens": 760518.0,
      "step": 82
    },
    {
      "epoch": 0.38694638694638694,
      "grad_norm": 2.407011283493592,
      "learning_rate": 8.019066368888222e-06,
      "loss": 0.4843,
      "mean_token_accuracy": 0.870231568813324,
      "num_tokens": 769855.0,
      "step": 83
    },
    {
      "epoch": 0.3916083916083916,
      "grad_norm": 2.72264066945379,
      "learning_rate": 7.958038330580067e-06,
      "loss": 0.5248,
      "mean_token_accuracy": 0.8568205535411835,
      "num_tokens": 780309.0,
      "step": 84
    },
    {
      "epoch": 0.3962703962703963,
      "grad_norm": 2.4515654207303332,
      "learning_rate": 7.89635901759967e-06,
      "loss": 0.4773,
      "mean_token_accuracy": 0.8676798641681671,
      "num_tokens": 790560.0,
      "step": 85
    },
    {
      "epoch": 0.40093240093240096,
      "grad_norm": 2.324209463411181,
      "learning_rate": 7.834044772318033e-06,
      "loss": 0.4696,
      "mean_token_accuracy": 0.8676293790340424,
      "num_tokens": 799972.0,
      "step": 86
    },
    {
      "epoch": 0.40559440559440557,
      "grad_norm": 2.4053156103185707,
      "learning_rate": 7.77111210533597e-06,
      "loss": 0.4985,
      "mean_token_accuracy": 0.8621641099452972,
      "num_tokens": 808998.0,
      "step": 87
    },
    {
      "epoch": 0.41025641025641024,
      "grad_norm": 2.520429601720593,
      "learning_rate": 7.707577691109519e-06,
      "loss": 0.5451,
      "mean_token_accuracy": 0.8532876372337341,
      "num_tokens": 818322.0,
      "step": 88
    },
    {
      "epoch": 0.4149184149184149,
      "grad_norm": 2.3400797394939,
      "learning_rate": 7.6434583635319e-06,
      "loss": 0.5222,
      "mean_token_accuracy": 0.8569265902042389,
      "num_tokens": 828468.0,
      "step": 89
    },
    {
      "epoch": 0.4195804195804196,
      "grad_norm": 2.3525161266543506,
      "learning_rate": 7.578771111473276e-06,
      "loss": 0.5352,
      "mean_token_accuracy": 0.8604429662227631,
      "num_tokens": 836909.0,
      "step": 90
    },
    {
      "epoch": 0.42424242424242425,
      "grad_norm": 2.2935148218456,
      "learning_rate": 7.513533074279427e-06,
      "loss": 0.4879,
      "mean_token_accuracy": 0.8654824793338776,
      "num_tokens": 845651.0,
      "step": 91
    },
    {
      "epoch": 0.4289044289044289,
      "grad_norm": 2.5471532688977283,
      "learning_rate": 7.4477615372305545e-06,
      "loss": 0.5336,
      "mean_token_accuracy": 0.8556940853595734,
      "num_tokens": 854858.0,
      "step": 92
    },
    {
      "epoch": 0.43356643356643354,
      "grad_norm": 2.3196850637517596,
      "learning_rate": 7.3814739269614265e-06,
      "loss": 0.4961,
      "mean_token_accuracy": 0.8647399842739105,
      "num_tokens": 863666.0,
      "step": 93
    },
    {
      "epoch": 0.4382284382284382,
      "grad_norm": 2.347290187249456,
      "learning_rate": 7.314687806844067e-06,
      "loss": 0.4886,
      "mean_token_accuracy": 0.8647028505802155,
      "num_tokens": 873307.0,
      "step": 94
    },
    {
      "epoch": 0.4428904428904429,
      "grad_norm": 2.2978535803132463,
      "learning_rate": 7.247420872334221e-06,
      "loss": 0.5159,
      "mean_token_accuracy": 0.8561632037162781,
      "num_tokens": 882337.0,
      "step": 95
    },
    {
      "epoch": 0.44755244755244755,
      "grad_norm": 2.2774719947241446,
      "learning_rate": 7.179690946282808e-06,
      "loss": 0.5066,
      "mean_token_accuracy": 0.8596616089344025,
      "num_tokens": 891528.0,
      "step": 96
    },
    {
      "epoch": 0.4522144522144522,
      "grad_norm": 2.334129070588941,
      "learning_rate": 7.111515974213639e-06,
      "loss": 0.4908,
      "mean_token_accuracy": 0.8704478442668915,
      "num_tokens": 900782.0,
      "step": 97
    },
    {
      "epoch": 0.4568764568764569,
      "grad_norm": 2.293727169950069,
      "learning_rate": 7.042914019568621e-06,
      "loss": 0.5212,
      "mean_token_accuracy": 0.8585403263568878,
      "num_tokens": 909811.0,
      "step": 98
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 2.1851004012786155,
      "learning_rate": 6.973903258921719e-06,
      "loss": 0.4606,
      "mean_token_accuracy": 0.8743224740028381,
      "num_tokens": 919377.0,
      "step": 99
    },
    {
      "epoch": 0.4662004662004662,
      "grad_norm": 2.325923644586097,
      "learning_rate": 6.904501977162949e-06,
      "loss": 0.5284,
      "mean_token_accuracy": 0.8553546667098999,
      "num_tokens": 929132.0,
      "step": 100
    },
    {
      "epoch": 0.47086247086247085,
      "grad_norm": 2.30316408858216,
      "learning_rate": 6.834728562653659e-06,
      "loss": 0.5216,
      "mean_token_accuracy": 0.8589284718036652,
      "num_tokens": 938942.0,
      "step": 101
    },
    {
      "epoch": 0.4755244755244755,
      "grad_norm": 2.277086694075995,
      "learning_rate": 6.764601502354403e-06,
      "loss": 0.5454,
      "mean_token_accuracy": 0.8548184633255005,
      "num_tokens": 948271.0,
      "step": 102
    },
    {
      "epoch": 0.4801864801864802,
      "grad_norm": 2.229671537868369,
      "learning_rate": 6.6941393769266995e-06,
      "loss": 0.4982,
      "mean_token_accuracy": 0.8633026480674744,
      "num_tokens": 957852.0,
      "step": 103
    },
    {
      "epoch": 0.48484848484848486,
      "grad_norm": 2.3287711507436546,
      "learning_rate": 6.6233608558099405e-06,
      "loss": 0.4835,
      "mean_token_accuracy": 0.8708520531654358,
      "num_tokens": 967455.0,
      "step": 104
    },
    {
      "epoch": 0.48951048951048953,
      "grad_norm": 2.3590620863308183,
      "learning_rate": 6.552284692274803e-06,
      "loss": 0.5271,
      "mean_token_accuracy": 0.8547607064247131,
      "num_tokens": 977978.0,
      "step": 105
    },
    {
      "epoch": 0.49417249417249415,
      "grad_norm": 2.4007057469187916,
      "learning_rate": 6.48092971845443e-06,
      "loss": 0.4918,
      "mean_token_accuracy": 0.8645793497562408,
      "num_tokens": 987423.0,
      "step": 106
    },
    {
      "epoch": 0.4988344988344988,
      "grad_norm": 2.4631312633850024,
      "learning_rate": 6.409314840354724e-06,
      "loss": 0.5325,
      "mean_token_accuracy": 0.8552180528640747,
      "num_tokens": 996656.0,
      "step": 107
    },
    {
      "epoch": 0.5034965034965035,
      "grad_norm": 2.4910422631378495,
      "learning_rate": 6.337459032845068e-06,
      "loss": 0.5376,
      "mean_token_accuracy": 0.8569700121879578,
      "num_tokens": 1005952.0,
      "step": 108
    },
    {
      "epoch": 0.5081585081585082,
      "grad_norm": 2.4102272333499637,
      "learning_rate": 6.2653813346308e-06,
      "loss": 0.507,
      "mean_token_accuracy": 0.8648414313793182,
      "num_tokens": 1015020.0,
      "step": 109
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 2.51736810310206,
      "learning_rate": 6.193100843208772e-06,
      "loss": 0.523,
      "mean_token_accuracy": 0.858431875705719,
      "num_tokens": 1023815.0,
      "step": 110
    },
    {
      "epoch": 0.5174825174825175,
      "grad_norm": 2.483927792491334,
      "learning_rate": 6.120636709807334e-06,
      "loss": 0.5181,
      "mean_token_accuracy": 0.8595193028450012,
      "num_tokens": 1032375.0,
      "step": 111
    },
    {
      "epoch": 0.5221445221445221,
      "grad_norm": 2.323963414867168,
      "learning_rate": 6.048008134312078e-06,
      "loss": 0.5038,
      "mean_token_accuracy": 0.8612522780895233,
      "num_tokens": 1041287.0,
      "step": 112
    },
    {
      "epoch": 0.5268065268065268,
      "grad_norm": 2.276747105471765,
      "learning_rate": 5.975234360178698e-06,
      "loss": 0.5316,
      "mean_token_accuracy": 0.8504720032215118,
      "num_tokens": 1052277.0,
      "step": 113
    },
    {
      "epoch": 0.5314685314685315,
      "grad_norm": 2.3768143861528803,
      "learning_rate": 5.902334669334287e-06,
      "loss": 0.5386,
      "mean_token_accuracy": 0.8551481366157532,
      "num_tokens": 1061624.0,
      "step": 114
    },
    {
      "epoch": 0.5361305361305362,
      "grad_norm": 2.4165115581800736,
      "learning_rate": 5.829328377068476e-06,
      "loss": 0.5038,
      "mean_token_accuracy": 0.8618453145027161,
      "num_tokens": 1070577.0,
      "step": 115
    },
    {
      "epoch": 0.5407925407925408,
      "grad_norm": 2.248719282657741,
      "learning_rate": 5.756234826915686e-06,
      "loss": 0.4713,
      "mean_token_accuracy": 0.8681885600090027,
      "num_tokens": 1080819.0,
      "step": 116
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 2.469628095969479,
      "learning_rate": 5.683073385529938e-06,
      "loss": 0.547,
      "mean_token_accuracy": 0.8538036346435547,
      "num_tokens": 1089899.0,
      "step": 117
    },
    {
      "epoch": 0.5501165501165501,
      "grad_norm": 2.2615322633589248,
      "learning_rate": 5.60986343755352e-06,
      "loss": 0.5377,
      "mean_token_accuracy": 0.8561292290687561,
      "num_tokens": 1099422.0,
      "step": 118
    },
    {
      "epoch": 0.5547785547785548,
      "grad_norm": 2.32979977164617,
      "learning_rate": 5.536624380480878e-06,
      "loss": 0.5034,
      "mean_token_accuracy": 0.8641898036003113,
      "num_tokens": 1108059.0,
      "step": 119
    },
    {
      "epoch": 0.5594405594405595,
      "grad_norm": 2.144339147442312,
      "learning_rate": 5.4633756195191235e-06,
      "loss": 0.485,
      "mean_token_accuracy": 0.870248019695282,
      "num_tokens": 1117344.0,
      "step": 120
    },
    {
      "epoch": 0.5641025641025641,
      "grad_norm": 2.011386160487392,
      "learning_rate": 5.390136562446482e-06,
      "loss": 0.4382,
      "mean_token_accuracy": 0.8803732097148895,
      "num_tokens": 1126493.0,
      "step": 121
    },
    {
      "epoch": 0.5687645687645687,
      "grad_norm": 2.286718045896851,
      "learning_rate": 5.316926614470063e-06,
      "loss": 0.512,
      "mean_token_accuracy": 0.8614423274993896,
      "num_tokens": 1135454.0,
      "step": 122
    },
    {
      "epoch": 0.5734265734265734,
      "grad_norm": 2.537037105358323,
      "learning_rate": 5.2437651730843165e-06,
      "loss": 0.5284,
      "mean_token_accuracy": 0.8529851138591766,
      "num_tokens": 1145120.0,
      "step": 123
    },
    {
      "epoch": 0.578088578088578,
      "grad_norm": 2.3343300964068625,
      "learning_rate": 5.170671622931527e-06,
      "loss": 0.4929,
      "mean_token_accuracy": 0.8647054135799408,
      "num_tokens": 1154569.0,
      "step": 124
    },
    {
      "epoch": 0.5827505827505828,
      "grad_norm": 2.097642279394774,
      "learning_rate": 5.097665330665714e-06,
      "loss": 0.4827,
      "mean_token_accuracy": 0.8696144223213196,
      "num_tokens": 1163964.0,
      "step": 125
    },
    {
      "epoch": 0.5874125874125874,
      "grad_norm": 2.398628734873881,
      "learning_rate": 5.024765639821305e-06,
      "loss": 0.5155,
      "mean_token_accuracy": 0.8605919480323792,
      "num_tokens": 1172705.0,
      "step": 126
    },
    {
      "epoch": 0.5920745920745921,
      "grad_norm": 2.0632119490542307,
      "learning_rate": 4.951991865687923e-06,
      "loss": 0.4514,
      "mean_token_accuracy": 0.872063010931015,
      "num_tokens": 1182428.0,
      "step": 127
    },
    {
      "epoch": 0.5967365967365967,
      "grad_norm": 2.2492976140784595,
      "learning_rate": 4.879363290192667e-06,
      "loss": 0.4992,
      "mean_token_accuracy": 0.8648842573165894,
      "num_tokens": 1191692.0,
      "step": 128
    },
    {
      "epoch": 0.6013986013986014,
      "grad_norm": 2.4763737198121976,
      "learning_rate": 4.806899156791231e-06,
      "loss": 0.5611,
      "mean_token_accuracy": 0.8499232232570648,
      "num_tokens": 1200804.0,
      "step": 129
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 2.3126829795227586,
      "learning_rate": 4.734618665369202e-06,
      "loss": 0.4903,
      "mean_token_accuracy": 0.8685189485549927,
      "num_tokens": 1209500.0,
      "step": 130
    },
    {
      "epoch": 0.6107226107226107,
      "grad_norm": 2.273459277732645,
      "learning_rate": 4.662540967154934e-06,
      "loss": 0.5257,
      "mean_token_accuracy": 0.8602505624294281,
      "num_tokens": 1218360.0,
      "step": 131
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 2.2496659663705834,
      "learning_rate": 4.5906851596452765e-06,
      "loss": 0.4955,
      "mean_token_accuracy": 0.8618595600128174,
      "num_tokens": 1227995.0,
      "step": 132
    },
    {
      "epoch": 0.62004662004662,
      "grad_norm": 2.259941152926795,
      "learning_rate": 4.519070281545571e-06,
      "loss": 0.5093,
      "mean_token_accuracy": 0.8600056171417236,
      "num_tokens": 1237369.0,
      "step": 133
    },
    {
      "epoch": 0.6247086247086248,
      "grad_norm": 2.439872298070353,
      "learning_rate": 4.447715307725197e-06,
      "loss": 0.5109,
      "mean_token_accuracy": 0.8662848472595215,
      "num_tokens": 1246811.0,
      "step": 134
    },
    {
      "epoch": 0.6293706293706294,
      "grad_norm": 2.3217691889727323,
      "learning_rate": 4.376639144190061e-06,
      "loss": 0.5308,
      "mean_token_accuracy": 0.856608659029007,
      "num_tokens": 1257728.0,
      "step": 135
    },
    {
      "epoch": 0.634032634032634,
      "grad_norm": 2.5154363333312952,
      "learning_rate": 4.305860623073304e-06,
      "loss": 0.5282,
      "mean_token_accuracy": 0.8532958924770355,
      "num_tokens": 1267082.0,
      "step": 136
    },
    {
      "epoch": 0.6386946386946387,
      "grad_norm": 2.150749527440618,
      "learning_rate": 4.2353984976456e-06,
      "loss": 0.4847,
      "mean_token_accuracy": 0.8674114644527435,
      "num_tokens": 1276457.0,
      "step": 137
    },
    {
      "epoch": 0.6433566433566433,
      "grad_norm": 2.4004488123223395,
      "learning_rate": 4.1652714373463435e-06,
      "loss": 0.5119,
      "mean_token_accuracy": 0.8605488240718842,
      "num_tokens": 1285816.0,
      "step": 138
    },
    {
      "epoch": 0.6480186480186481,
      "grad_norm": 2.309626827806937,
      "learning_rate": 4.095498022837051e-06,
      "loss": 0.4974,
      "mean_token_accuracy": 0.864313542842865,
      "num_tokens": 1294833.0,
      "step": 139
    },
    {
      "epoch": 0.6526806526806527,
      "grad_norm": 2.3480241095920302,
      "learning_rate": 4.026096741078281e-06,
      "loss": 0.4916,
      "mean_token_accuracy": 0.8628659248352051,
      "num_tokens": 1304626.0,
      "step": 140
    },
    {
      "epoch": 0.6573426573426573,
      "grad_norm": 2.30703663882747,
      "learning_rate": 3.957085980431382e-06,
      "loss": 0.505,
      "mean_token_accuracy": 0.8645538091659546,
      "num_tokens": 1314352.0,
      "step": 141
    },
    {
      "epoch": 0.662004662004662,
      "grad_norm": 2.3462321603678236,
      "learning_rate": 3.888484025786364e-06,
      "loss": 0.5181,
      "mean_token_accuracy": 0.8561761677265167,
      "num_tokens": 1323815.0,
      "step": 142
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 2.1292499679236774,
      "learning_rate": 3.820309053717195e-06,
      "loss": 0.5081,
      "mean_token_accuracy": 0.8614768087863922,
      "num_tokens": 1333085.0,
      "step": 143
    },
    {
      "epoch": 0.6713286713286714,
      "grad_norm": 2.378840264366356,
      "learning_rate": 3.75257912766578e-06,
      "loss": 0.557,
      "mean_token_accuracy": 0.8496690094470978,
      "num_tokens": 1342490.0,
      "step": 144
    },
    {
      "epoch": 0.675990675990676,
      "grad_norm": 2.4574932080087577,
      "learning_rate": 3.6853121931559334e-06,
      "loss": 0.5412,
      "mean_token_accuracy": 0.8501175343990326,
      "num_tokens": 1351579.0,
      "step": 145
    },
    {
      "epoch": 0.6806526806526807,
      "grad_norm": 2.174435905448035,
      "learning_rate": 3.618526073038574e-06,
      "loss": 0.5034,
      "mean_token_accuracy": 0.8580139875411987,
      "num_tokens": 1363017.0,
      "step": 146
    },
    {
      "epoch": 0.6853146853146853,
      "grad_norm": 2.3281070276733042,
      "learning_rate": 3.552238462769446e-06,
      "loss": 0.4875,
      "mean_token_accuracy": 0.864938497543335,
      "num_tokens": 1372276.0,
      "step": 147
    },
    {
      "epoch": 0.6899766899766899,
      "grad_norm": 2.21912732761688,
      "learning_rate": 3.4864669257205745e-06,
      "loss": 0.497,
      "mean_token_accuracy": 0.8623648881912231,
      "num_tokens": 1381474.0,
      "step": 148
    },
    {
      "epoch": 0.6946386946386947,
      "grad_norm": 2.381930397952038,
      "learning_rate": 3.4212288885267246e-06,
      "loss": 0.5307,
      "mean_token_accuracy": 0.8609063029289246,
      "num_tokens": 1390844.0,
      "step": 149
    },
    {
      "epoch": 0.6993006993006993,
      "grad_norm": 2.2973850752876053,
      "learning_rate": 3.3565416364681016e-06,
      "loss": 0.4724,
      "mean_token_accuracy": 0.8680506646633148,
      "num_tokens": 1399649.0,
      "step": 150
    },
    {
      "epoch": 0.703962703962704,
      "grad_norm": 2.4752144217262906,
      "learning_rate": 3.2924223088904816e-06,
      "loss": 0.5534,
      "mean_token_accuracy": 0.8529053032398224,
      "num_tokens": 1409096.0,
      "step": 151
    },
    {
      "epoch": 0.7086247086247086,
      "grad_norm": 2.4697356748705404,
      "learning_rate": 3.228887894664029e-06,
      "loss": 0.5833,
      "mean_token_accuracy": 0.8456442058086395,
      "num_tokens": 1417944.0,
      "step": 152
    },
    {
      "epoch": 0.7132867132867133,
      "grad_norm": 2.2561615402124398,
      "learning_rate": 3.1659552276819693e-06,
      "loss": 0.5316,
      "mean_token_accuracy": 0.8573636114597321,
      "num_tokens": 1427079.0,
      "step": 153
    },
    {
      "epoch": 0.717948717948718,
      "grad_norm": 2.2330254659734594,
      "learning_rate": 3.1036409824003324e-06,
      "loss": 0.5144,
      "mean_token_accuracy": 0.861716091632843,
      "num_tokens": 1436521.0,
      "step": 154
    },
    {
      "epoch": 0.7226107226107226,
      "grad_norm": 2.2412294204365333,
      "learning_rate": 3.0419616694199327e-06,
      "loss": 0.5027,
      "mean_token_accuracy": 0.8678925633430481,
      "num_tokens": 1445553.0,
      "step": 155
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 2.259593929352767,
      "learning_rate": 2.98093363111178e-06,
      "loss": 0.5135,
      "mean_token_accuracy": 0.8669766783714294,
      "num_tokens": 1453774.0,
      "step": 156
    },
    {
      "epoch": 0.7319347319347319,
      "grad_norm": 2.4594381420216607,
      "learning_rate": 2.92057303728704e-06,
      "loss": 0.539,
      "mean_token_accuracy": 0.855047732591629,
      "num_tokens": 1462611.0,
      "step": 157
    },
    {
      "epoch": 0.7365967365967366,
      "grad_norm": 2.316483487426212,
      "learning_rate": 2.860895880912735e-06,
      "loss": 0.4671,
      "mean_token_accuracy": 0.8716062903404236,
      "num_tokens": 1471736.0,
      "step": 158
    },
    {
      "epoch": 0.7412587412587412,
      "grad_norm": 2.3177493374035256,
      "learning_rate": 2.801917973874294e-06,
      "loss": 0.5158,
      "mean_token_accuracy": 0.8569830656051636,
      "num_tokens": 1482117.0,
      "step": 159
    },
    {
      "epoch": 0.745920745920746,
      "grad_norm": 2.3971972763371077,
      "learning_rate": 2.7436549427860766e-06,
      "loss": 0.5361,
      "mean_token_accuracy": 0.8565348982810974,
      "num_tokens": 1491292.0,
      "step": 160
    },
    {
      "epoch": 0.7505827505827506,
      "grad_norm": 2.1411052658171426,
      "learning_rate": 2.6861222248509926e-06,
      "loss": 0.4865,
      "mean_token_accuracy": 0.8685896992683411,
      "num_tokens": 1500735.0,
      "step": 161
    },
    {
      "epoch": 0.7552447552447552,
      "grad_norm": 2.276293848116446,
      "learning_rate": 2.6293350637703123e-06,
      "loss": 0.4799,
      "mean_token_accuracy": 0.8709887862205505,
      "num_tokens": 1510190.0,
      "step": 162
    },
    {
      "epoch": 0.7599067599067599,
      "grad_norm": 2.4261320460180955,
      "learning_rate": 2.5733085057047325e-06,
      "loss": 0.5242,
      "mean_token_accuracy": 0.857810914516449,
      "num_tokens": 1519207.0,
      "step": 163
    },
    {
      "epoch": 0.7645687645687645,
      "grad_norm": 2.324362641758302,
      "learning_rate": 2.518057395287792e-06,
      "loss": 0.4984,
      "mean_token_accuracy": 0.8615204393863678,
      "num_tokens": 1528808.0,
      "step": 164
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 2.3529932795802018,
      "learning_rate": 2.463596371692681e-06,
      "loss": 0.4868,
      "mean_token_accuracy": 0.867795467376709,
      "num_tokens": 1537765.0,
      "step": 165
    },
    {
      "epoch": 0.7738927738927739,
      "grad_norm": 2.419146597375076,
      "learning_rate": 2.409939864753487e-06,
      "loss": 0.505,
      "mean_token_accuracy": 0.8625264167785645,
      "num_tokens": 1547822.0,
      "step": 166
    },
    {
      "epoch": 0.7785547785547785,
      "grad_norm": 2.116381511893311,
      "learning_rate": 2.3571020911419067e-06,
      "loss": 0.463,
      "mean_token_accuracy": 0.8741449117660522,
      "num_tokens": 1557620.0,
      "step": 167
    },
    {
      "epoch": 0.7832167832167832,
      "grad_norm": 2.3300308626713195,
      "learning_rate": 2.3050970506004463e-06,
      "loss": 0.4853,
      "mean_token_accuracy": 0.8641858994960785,
      "num_tokens": 1567111.0,
      "step": 168
    },
    {
      "epoch": 0.7878787878787878,
      "grad_norm": 2.2767831001273713,
      "learning_rate": 2.2539385222330797e-06,
      "loss": 0.4728,
      "mean_token_accuracy": 0.8725703656673431,
      "num_tokens": 1575836.0,
      "step": 169
    },
    {
      "epoch": 0.7925407925407926,
      "grad_norm": 2.334989006961217,
      "learning_rate": 2.203640060854387e-06,
      "loss": 0.504,
      "mean_token_accuracy": 0.8669374287128448,
      "num_tokens": 1584822.0,
      "step": 170
    },
    {
      "epoch": 0.7972027972027972,
      "grad_norm": 2.2738559617797303,
      "learning_rate": 2.1542149933981014e-06,
      "loss": 0.5421,
      "mean_token_accuracy": 0.8533051311969757,
      "num_tokens": 1594428.0,
      "step": 171
    },
    {
      "epoch": 0.8018648018648019,
      "grad_norm": 2.312827620699107,
      "learning_rate": 2.10567641538605e-06,
      "loss": 0.5151,
      "mean_token_accuracy": 0.8604055345058441,
      "num_tokens": 1603786.0,
      "step": 172
    },
    {
      "epoch": 0.8065268065268065,
      "grad_norm": 2.1729722956072837,
      "learning_rate": 2.058037187458398e-06,
      "loss": 0.4768,
      "mean_token_accuracy": 0.8706162571907043,
      "num_tokens": 1612747.0,
      "step": 173
    },
    {
      "epoch": 0.8111888111888111,
      "grad_norm": 2.1288452988785256,
      "learning_rate": 2.011309931966136e-06,
      "loss": 0.4729,
      "mean_token_accuracy": 0.8719466626644135,
      "num_tokens": 1621819.0,
      "step": 174
    },
    {
      "epoch": 0.8158508158508159,
      "grad_norm": 2.664000380251652,
      "learning_rate": 1.965507029626695e-06,
      "loss": 0.5164,
      "mean_token_accuracy": 0.8638178110122681,
      "num_tokens": 1631685.0,
      "step": 175
    },
    {
      "epoch": 0.8205128205128205,
      "grad_norm": 2.1258796541566043,
      "learning_rate": 1.920640616243589e-06,
      "loss": 0.4603,
      "mean_token_accuracy": 0.877100944519043,
      "num_tokens": 1640794.0,
      "step": 176
    },
    {
      "epoch": 0.8251748251748252,
      "grad_norm": 2.3499685441879845,
      "learning_rate": 1.8767225794909484e-06,
      "loss": 0.4905,
      "mean_token_accuracy": 0.8642638325691223,
      "num_tokens": 1650186.0,
      "step": 177
    },
    {
      "epoch": 0.8298368298368298,
      "grad_norm": 2.2575222389518976,
      "learning_rate": 1.8337645557637929e-06,
      "loss": 0.4977,
      "mean_token_accuracy": 0.8628556430339813,
      "num_tokens": 1659577.0,
      "step": 178
    },
    {
      "epoch": 0.8344988344988346,
      "grad_norm": 2.2749724550879358,
      "learning_rate": 1.7917779270948887e-06,
      "loss": 0.4978,
      "mean_token_accuracy": 0.8586052358150482,
      "num_tokens": 1669821.0,
      "step": 179
    },
    {
      "epoch": 0.8391608391608392,
      "grad_norm": 2.503216275614053,
      "learning_rate": 1.7507738181390027e-06,
      "loss": 0.542,
      "mean_token_accuracy": 0.8548881709575653,
      "num_tokens": 1679579.0,
      "step": 180
    },
    {
      "epoch": 0.8438228438228438,
      "grad_norm": 2.309496420384001,
      "learning_rate": 1.7107630932253383e-06,
      "loss": 0.4992,
      "mean_token_accuracy": 0.866202175617218,
      "num_tokens": 1688922.0,
      "step": 181
    },
    {
      "epoch": 0.8484848484848485,
      "grad_norm": 2.2088000893568274,
      "learning_rate": 1.6717563534789594e-06,
      "loss": 0.5042,
      "mean_token_accuracy": 0.8664792478084564,
      "num_tokens": 1697949.0,
      "step": 182
    },
    {
      "epoch": 0.8531468531468531,
      "grad_norm": 2.1790077110948243,
      "learning_rate": 1.6337639340119476e-06,
      "loss": 0.5049,
      "mean_token_accuracy": 0.8635801076889038,
      "num_tokens": 1708365.0,
      "step": 183
    },
    {
      "epoch": 0.8578088578088578,
      "grad_norm": 2.322644087839244,
      "learning_rate": 1.596795901185037e-06,
      "loss": 0.5093,
      "mean_token_accuracy": 0.8596910536289215,
      "num_tokens": 1717340.0,
      "step": 184
    },
    {
      "epoch": 0.8624708624708625,
      "grad_norm": 2.153623703135592,
      "learning_rate": 1.5608620499404628e-06,
      "loss": 0.5136,
      "mean_token_accuracy": 0.8647525906562805,
      "num_tokens": 1727305.0,
      "step": 185
    },
    {
      "epoch": 0.8671328671328671,
      "grad_norm": 2.4051184830525516,
      "learning_rate": 1.5259719012067249e-06,
      "loss": 0.5403,
      "mean_token_accuracy": 0.8549070060253143,
      "num_tokens": 1737093.0,
      "step": 186
    },
    {
      "epoch": 0.8717948717948718,
      "grad_norm": 2.2034389449217735,
      "learning_rate": 1.4921346993759453e-06,
      "loss": 0.5071,
      "mean_token_accuracy": 0.8640574812889099,
      "num_tokens": 1746772.0,
      "step": 187
    },
    {
      "epoch": 0.8764568764568764,
      "grad_norm": 2.106806136374632,
      "learning_rate": 1.459359409854505e-06,
      "loss": 0.4552,
      "mean_token_accuracy": 0.8732733726501465,
      "num_tokens": 1756307.0,
      "step": 188
    },
    {
      "epoch": 0.8811188811188811,
      "grad_norm": 2.3203500819249596,
      "learning_rate": 1.4276547166875946e-06,
      "loss": 0.5126,
      "mean_token_accuracy": 0.8606568872928619,
      "num_tokens": 1766010.0,
      "step": 189
    },
    {
      "epoch": 0.8857808857808858,
      "grad_norm": 2.258227696938333,
      "learning_rate": 1.397029020258313e-06,
      "loss": 0.5263,
      "mean_token_accuracy": 0.8579614460468292,
      "num_tokens": 1774885.0,
      "step": 190
    },
    {
      "epoch": 0.8904428904428905,
      "grad_norm": 2.2257353197975673,
      "learning_rate": 1.367490435061928e-06,
      "loss": 0.4885,
      "mean_token_accuracy": 0.8630119860172272,
      "num_tokens": 1785248.0,
      "step": 191
    },
    {
      "epoch": 0.8951048951048951,
      "grad_norm": 2.1640943197597413,
      "learning_rate": 1.3390467875558855e-06,
      "loss": 0.4807,
      "mean_token_accuracy": 0.869134396314621,
      "num_tokens": 1794525.0,
      "step": 192
    },
    {
      "epoch": 0.8997668997668997,
      "grad_norm": 2.1326960666576587,
      "learning_rate": 1.3117056140861317e-06,
      "loss": 0.5064,
      "mean_token_accuracy": 0.8599035441875458,
      "num_tokens": 1804413.0,
      "step": 193
    },
    {
      "epoch": 0.9044289044289044,
      "grad_norm": 2.1347319106028224,
      "learning_rate": 1.285474158890304e-06,
      "loss": 0.4704,
      "mean_token_accuracy": 0.8728452920913696,
      "num_tokens": 1813695.0,
      "step": 194
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 2.020006943601936,
      "learning_rate": 1.2603593721783219e-06,
      "loss": 0.4503,
      "mean_token_accuracy": 0.8767442107200623,
      "num_tokens": 1823851.0,
      "step": 195
    },
    {
      "epoch": 0.9137529137529138,
      "grad_norm": 2.1574624570969947,
      "learning_rate": 1.2363679082908766e-06,
      "loss": 0.4805,
      "mean_token_accuracy": 0.8633581399917603,
      "num_tokens": 1833828.0,
      "step": 196
    },
    {
      "epoch": 0.9184149184149184,
      "grad_norm": 2.3395117245082657,
      "learning_rate": 1.2135061239363161e-06,
      "loss": 0.5083,
      "mean_token_accuracy": 0.863542228937149,
      "num_tokens": 1842998.0,
      "step": 197
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.28472593102768,
      "learning_rate": 1.1917800765063954e-06,
      "loss": 0.5163,
      "mean_token_accuracy": 0.8649951219558716,
      "num_tokens": 1851919.0,
      "step": 198
    },
    {
      "epoch": 0.9277389277389277,
      "grad_norm": 2.364467187406286,
      "learning_rate": 1.1711955224713209e-06,
      "loss": 0.5198,
      "mean_token_accuracy": 0.8597099781036377,
      "num_tokens": 1860906.0,
      "step": 199
    },
    {
      "epoch": 0.9324009324009324,
      "grad_norm": 2.3628488712143807,
      "learning_rate": 1.1517579158545386e-06,
      "loss": 0.5204,
      "mean_token_accuracy": 0.8572766482830048,
      "num_tokens": 1870470.0,
      "step": 200
    },
    {
      "epoch": 0.9370629370629371,
      "grad_norm": 2.3509216737104297,
      "learning_rate": 1.1334724067876463e-06,
      "loss": 0.5263,
      "mean_token_accuracy": 0.857895165681839,
      "num_tokens": 1879984.0,
      "step": 201
    },
    {
      "epoch": 0.9417249417249417,
      "grad_norm": 2.27625332892735,
      "learning_rate": 1.1163438401458358e-06,
      "loss": 0.4761,
      "mean_token_accuracy": 0.8712551891803741,
      "num_tokens": 1889348.0,
      "step": 202
    },
    {
      "epoch": 0.9463869463869464,
      "grad_norm": 2.2664430838438303,
      "learning_rate": 1.1003767542642021e-06,
      "loss": 0.518,
      "mean_token_accuracy": 0.8641975820064545,
      "num_tokens": 1898543.0,
      "step": 203
    },
    {
      "epoch": 0.951048951048951,
      "grad_norm": 2.154720441862823,
      "learning_rate": 1.0855753797352868e-06,
      "loss": 0.4834,
      "mean_token_accuracy": 0.8657752573490143,
      "num_tokens": 1908052.0,
      "step": 204
    },
    {
      "epoch": 0.9557109557109557,
      "grad_norm": 2.184478724115355,
      "learning_rate": 1.0719436382881466e-06,
      "loss": 0.4846,
      "mean_token_accuracy": 0.8683419525623322,
      "num_tokens": 1917592.0,
      "step": 205
    },
    {
      "epoch": 0.9603729603729604,
      "grad_norm": 2.218994175211714,
      "learning_rate": 1.0594851417492665e-06,
      "loss": 0.5136,
      "mean_token_accuracy": 0.8631373941898346,
      "num_tokens": 1927404.0,
      "step": 206
    },
    {
      "epoch": 0.965034965034965,
      "grad_norm": 2.38088407847355,
      "learning_rate": 1.0482031910855804e-06,
      "loss": 0.5129,
      "mean_token_accuracy": 0.857024222612381,
      "num_tokens": 1936273.0,
      "step": 207
    },
    {
      "epoch": 0.9696969696969697,
      "grad_norm": 2.2270852636684437,
      "learning_rate": 1.0381007755298547e-06,
      "loss": 0.4865,
      "mean_token_accuracy": 0.8681848645210266,
      "num_tokens": 1944830.0,
      "step": 208
    },
    {
      "epoch": 0.9743589743589743,
      "grad_norm": 2.269354458552895,
      "learning_rate": 1.029180571788672e-06,
      "loss": 0.5067,
      "mean_token_accuracy": 0.8632671236991882,
      "num_tokens": 1954306.0,
      "step": 209
    },
    {
      "epoch": 0.9790209790209791,
      "grad_norm": 2.1740483509672273,
      "learning_rate": 1.021444943333218e-06,
      "loss": 0.4961,
      "mean_token_accuracy": 0.8667227923870087,
      "num_tokens": 1964061.0,
      "step": 210
    },
    {
      "epoch": 0.9836829836829837,
      "grad_norm": 2.1697738914903746,
      "learning_rate": 1.0148959397730637e-06,
      "loss": 0.4679,
      "mean_token_accuracy": 0.8761565685272217,
      "num_tokens": 1973286.0,
      "step": 211
    },
    {
      "epoch": 0.9883449883449883,
      "grad_norm": 2.241781686365417,
      "learning_rate": 1.0095352963131057e-06,
      "loss": 0.4946,
      "mean_token_accuracy": 0.8629972040653229,
      "num_tokens": 1983373.0,
      "step": 212
    },
    {
      "epoch": 0.993006993006993,
      "grad_norm": 2.126258981095298,
      "learning_rate": 1.0053644332938118e-06,
      "loss": 0.4272,
      "mean_token_accuracy": 0.8793900609016418,
      "num_tokens": 1992207.0,
      "step": 213
    },
    {
      "epoch": 0.9976689976689976,
      "grad_norm": 2.3561589852941816,
      "learning_rate": 1.0023844558148912e-06,
      "loss": 0.4912,
      "mean_token_accuracy": 0.8664555549621582,
      "num_tokens": 2001400.0,
      "step": 214
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.3561589852941816,
      "learning_rate": 1.0005961534424925e-06,
      "loss": 0.467,
      "mean_token_accuracy": 0.8875675797462463,
      "num_tokens": 2003256.0,
      "step": 215
    },
    {
      "epoch": 1.0,
      "step": 215,
      "total_flos": 64885863989248.0,
      "train_loss": 0.5904712042143179,
      "train_runtime": 1997.1704,
      "train_samples_per_second": 3.43,
      "train_steps_per_second": 0.108
    }
  ],
  "logging_steps": 1,
  "max_steps": 215,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 64885863989248.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}