{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 142,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.78125,
      "epoch": 0.007042253521126761,
      "grad_norm": 5.375,
      "learning_rate": 5e-05,
      "loss": 2.342320680618286,
      "mean_token_accuracy": 0.5398398637771606,
      "num_tokens": 2631.0,
      "step": 1
    },
    {
      "entropy": 1.7265625,
      "epoch": 0.014084507042253521,
      "grad_norm": 4.28125,
      "learning_rate": 5e-05,
      "loss": 2.184455394744873,
      "mean_token_accuracy": 0.5707964897155762,
      "num_tokens": 5577.0,
      "step": 2
    },
    {
      "entropy": 1.5859375,
      "epoch": 0.02112676056338028,
      "grad_norm": 3.796875,
      "learning_rate": 5e-05,
      "loss": 1.9440301656723022,
      "mean_token_accuracy": 0.6223160624504089,
      "num_tokens": 8426.0,
      "step": 3
    },
    {
      "entropy": 1.75,
      "epoch": 0.028169014084507043,
      "grad_norm": 3.234375,
      "learning_rate": 5e-05,
      "loss": 2.034520149230957,
      "mean_token_accuracy": 0.5993841886520386,
      "num_tokens": 11357.0,
      "step": 4
    },
    {
      "entropy": 1.8984375,
      "epoch": 0.035211267605633804,
      "grad_norm": 3.3125,
      "learning_rate": 5e-05,
      "loss": 2.151127338409424,
      "mean_token_accuracy": 0.563842236995697,
      "num_tokens": 13926.0,
      "step": 5
    },
    {
      "entropy": 1.890625,
      "epoch": 0.04225352112676056,
      "grad_norm": 2.546875,
      "learning_rate": 5e-05,
      "loss": 2.066777229309082,
      "mean_token_accuracy": 0.5754261612892151,
      "num_tokens": 17043.0,
      "step": 6
    },
    {
      "entropy": 2.03125,
      "epoch": 0.04929577464788732,
      "grad_norm": 2.9375,
      "learning_rate": 5e-05,
      "loss": 2.2290213108062744,
      "mean_token_accuracy": 0.5708707571029663,
      "num_tokens": 19612.0,
      "step": 7
    },
    {
      "entropy": 1.8046875,
      "epoch": 0.056338028169014086,
      "grad_norm": 2.59375,
      "learning_rate": 5e-05,
      "loss": 1.9348505735397339,
      "mean_token_accuracy": 0.6066350936889648,
      "num_tokens": 22363.0,
      "step": 8
    },
    {
      "entropy": 1.7734375,
      "epoch": 0.06338028169014084,
      "grad_norm": 2.4375,
      "learning_rate": 5e-05,
      "loss": 1.8786349296569824,
      "mean_token_accuracy": 0.628695011138916,
      "num_tokens": 25145.0,
      "step": 9
    },
    {
      "entropy": 1.8515625,
      "epoch": 0.07042253521126761,
      "grad_norm": 2.296875,
      "learning_rate": 5e-05,
      "loss": 1.9190694093704224,
      "mean_token_accuracy": 0.6125850081443787,
      "num_tokens": 28093.0,
      "step": 10
    },
    {
      "entropy": 1.8515625,
      "epoch": 0.07746478873239436,
      "grad_norm": 2.171875,
      "learning_rate": 5e-05,
      "loss": 1.891284704208374,
      "mean_token_accuracy": 0.6322028040885925,
      "num_tokens": 31138.0,
      "step": 11
    },
    {
      "entropy": 1.953125,
      "epoch": 0.08450704225352113,
      "grad_norm": 2.421875,
      "learning_rate": 5e-05,
      "loss": 1.948248267173767,
      "mean_token_accuracy": 0.6148879528045654,
      "num_tokens": 33779.0,
      "step": 12
    },
    {
      "entropy": 1.9609375,
      "epoch": 0.09154929577464789,
      "grad_norm": 2.671875,
      "learning_rate": 5e-05,
      "loss": 1.968241810798645,
      "mean_token_accuracy": 0.6172575950622559,
      "num_tokens": 36128.0,
      "step": 13
    },
    {
      "entropy": 1.8828125,
      "epoch": 0.09859154929577464,
      "grad_norm": 2.015625,
      "learning_rate": 5e-05,
      "loss": 1.902205228805542,
      "mean_token_accuracy": 0.6156116724014282,
      "num_tokens": 39185.0,
      "step": 14
    },
    {
      "entropy": 1.703125,
      "epoch": 0.1056338028169014,
      "grad_norm": 2.328125,
      "learning_rate": 5e-05,
      "loss": 1.6213911771774292,
      "mean_token_accuracy": 0.6750187277793884,
      "num_tokens": 41867.0,
      "step": 15
    },
    {
      "entropy": 2.015625,
      "epoch": 0.11267605633802817,
      "grad_norm": 2.234375,
      "learning_rate": 5e-05,
      "loss": 1.9176678657531738,
      "mean_token_accuracy": 0.6337976455688477,
      "num_tokens": 44603.0,
      "step": 16
    },
    {
      "entropy": 1.921875,
      "epoch": 0.11971830985915492,
      "grad_norm": 1.921875,
      "learning_rate": 5e-05,
      "loss": 1.8336955308914185,
      "mean_token_accuracy": 0.6507288813591003,
      "num_tokens": 48041.0,
      "step": 17
    },
    {
      "entropy": 1.828125,
      "epoch": 0.1267605633802817,
      "grad_norm": 2.046875,
      "learning_rate": 5e-05,
      "loss": 1.7068480253219604,
      "mean_token_accuracy": 0.6735284924507141,
      "num_tokens": 51158.0,
      "step": 18
    },
    {
      "entropy": 1.7421875,
      "epoch": 0.13380281690140844,
      "grad_norm": 1.8984375,
      "learning_rate": 5e-05,
      "loss": 1.6770858764648438,
      "mean_token_accuracy": 0.6662946343421936,
      "num_tokens": 54750.0,
      "step": 19
    },
    {
      "entropy": 2.109375,
      "epoch": 0.14084507042253522,
      "grad_norm": 2.828125,
      "learning_rate": 5e-05,
      "loss": 1.9946776628494263,
      "mean_token_accuracy": 0.6195476651191711,
      "num_tokens": 57234.0,
      "step": 20
    },
    {
      "entropy": 1.6328125,
      "epoch": 0.14788732394366197,
      "grad_norm": 1.90625,
      "learning_rate": 5e-05,
      "loss": 1.5617338418960571,
      "mean_token_accuracy": 0.6951566934585571,
      "num_tokens": 60752.0,
      "step": 21
    },
    {
      "entropy": 1.78125,
      "epoch": 0.15492957746478872,
      "grad_norm": 2.515625,
      "learning_rate": 5e-05,
      "loss": 1.7242701053619385,
      "mean_token_accuracy": 0.6592110395431519,
      "num_tokens": 63219.0,
      "step": 22
    },
    {
      "entropy": 1.6796875,
      "epoch": 0.1619718309859155,
      "grad_norm": 1.9375,
      "learning_rate": 5e-05,
      "loss": 1.6089009046554565,
      "mean_token_accuracy": 0.6697189807891846,
      "num_tokens": 66394.0,
      "step": 23
    },
    {
      "entropy": 1.9296875,
      "epoch": 0.16901408450704225,
      "grad_norm": 2.125,
      "learning_rate": 5e-05,
      "loss": 1.8391156196594238,
      "mean_token_accuracy": 0.6323254704475403,
      "num_tokens": 69081.0,
      "step": 24
    },
    {
      "entropy": 1.7265625,
      "epoch": 0.176056338028169,
      "grad_norm": 2.15625,
      "learning_rate": 5e-05,
      "loss": 1.6465730667114258,
      "mean_token_accuracy": 0.654959499835968,
      "num_tokens": 71680.0,
      "step": 25
    },
    {
      "entropy": 1.9609375,
      "epoch": 0.18309859154929578,
      "grad_norm": 2.03125,
      "learning_rate": 5e-05,
      "loss": 1.873896598815918,
      "mean_token_accuracy": 0.6338028311729431,
      "num_tokens": 74670.0,
      "step": 26
    },
    {
      "entropy": 1.6875,
      "epoch": 0.19014084507042253,
      "grad_norm": 1.84375,
      "learning_rate": 5e-05,
      "loss": 1.629098653793335,
      "mean_token_accuracy": 0.6646864414215088,
      "num_tokens": 77708.0,
      "step": 27
    },
    {
      "entropy": 1.609375,
      "epoch": 0.19718309859154928,
      "grad_norm": 1.8671875,
      "learning_rate": 5e-05,
      "loss": 1.5465565919876099,
      "mean_token_accuracy": 0.6922129988670349,
      "num_tokens": 80965.0,
      "step": 28
    },
    {
      "entropy": 1.59375,
      "epoch": 0.20422535211267606,
      "grad_norm": 2.140625,
      "learning_rate": 5e-05,
      "loss": 1.5382331609725952,
      "mean_token_accuracy": 0.6933178901672363,
      "num_tokens": 83562.0,
      "step": 29
    },
    {
      "entropy": 1.5390625,
      "epoch": 0.2112676056338028,
      "grad_norm": 1.9921875,
      "learning_rate": 5e-05,
      "loss": 1.4340285062789917,
      "mean_token_accuracy": 0.6944348216056824,
      "num_tokens": 86463.0,
      "step": 30
    },
    {
      "entropy": 1.5,
      "epoch": 0.21830985915492956,
      "grad_norm": 1.953125,
      "learning_rate": 5e-05,
      "loss": 1.41927969455719,
      "mean_token_accuracy": 0.7242607474327087,
      "num_tokens": 89278.0,
      "step": 31
    },
    {
      "entropy": 1.484375,
      "epoch": 0.22535211267605634,
      "grad_norm": 2.328125,
      "learning_rate": 5e-05,
      "loss": 1.395577311515808,
      "mean_token_accuracy": 0.7233359813690186,
      "num_tokens": 91780.0,
      "step": 32
    },
    {
      "entropy": 1.5234375,
      "epoch": 0.2323943661971831,
      "grad_norm": 1.8671875,
      "learning_rate": 5e-05,
      "loss": 1.4273605346679688,
      "mean_token_accuracy": 0.7213010191917419,
      "num_tokens": 94924.0,
      "step": 33
    },
    {
      "entropy": 1.640625,
      "epoch": 0.23943661971830985,
      "grad_norm": 1.9375,
      "learning_rate": 5e-05,
      "loss": 1.549876093864441,
      "mean_token_accuracy": 0.6991525292396545,
      "num_tokens": 98236.0,
      "step": 34
    },
    {
      "entropy": 1.3984375,
      "epoch": 0.24647887323943662,
      "grad_norm": 2.5625,
      "learning_rate": 5e-05,
      "loss": 1.2571226358413696,
      "mean_token_accuracy": 0.7764654159545898,
      "num_tokens": 100530.0,
      "step": 35
    },
    {
      "entropy": 1.484375,
      "epoch": 0.2535211267605634,
      "grad_norm": 1.9921875,
      "learning_rate": 5e-05,
      "loss": 1.430403709411621,
      "mean_token_accuracy": 0.7238532304763794,
      "num_tokens": 103808.0,
      "step": 36
    },
    {
      "entropy": 1.4375,
      "epoch": 0.2605633802816901,
      "grad_norm": 1.9296875,
      "learning_rate": 5e-05,
      "loss": 1.370211124420166,
      "mean_token_accuracy": 0.7603058218955994,
      "num_tokens": 106824.0,
      "step": 37
    },
    {
      "entropy": 1.3046875,
      "epoch": 0.2676056338028169,
      "grad_norm": 1.9921875,
      "learning_rate": 5e-05,
      "loss": 1.160237431526184,
      "mean_token_accuracy": 0.7812739610671997,
      "num_tokens": 109438.0,
      "step": 38
    },
    {
      "entropy": 1.40625,
      "epoch": 0.2746478873239437,
      "grad_norm": 2.265625,
      "learning_rate": 5e-05,
      "loss": 1.3372321128845215,
      "mean_token_accuracy": 0.7477554678916931,
      "num_tokens": 111785.0,
      "step": 39
    },
    {
      "entropy": 1.3671875,
      "epoch": 0.28169014084507044,
      "grad_norm": 1.84375,
      "learning_rate": 5e-05,
      "loss": 1.3203842639923096,
      "mean_token_accuracy": 0.7538297176361084,
      "num_tokens": 114600.0,
      "step": 40
    },
    {
      "entropy": 1.21875,
      "epoch": 0.2887323943661972,
      "grad_norm": 2.203125,
      "learning_rate": 5e-05,
      "loss": 1.1055949926376343,
      "mean_token_accuracy": 0.7860545516014099,
      "num_tokens": 116917.0,
      "step": 41
    },
    {
      "entropy": 1.484375,
      "epoch": 0.29577464788732394,
      "grad_norm": 1.546875,
      "learning_rate": 5e-05,
      "loss": 1.403846025466919,
      "mean_token_accuracy": 0.7451103925704956,
      "num_tokens": 120095.0,
      "step": 42
    },
    {
      "entropy": 1.328125,
      "epoch": 0.3028169014084507,
      "grad_norm": 1.65625,
      "learning_rate": 5e-05,
      "loss": 1.3071074485778809,
      "mean_token_accuracy": 0.7589316964149475,
      "num_tokens": 122986.0,
      "step": 43
    },
    {
      "entropy": 1.3203125,
      "epoch": 0.30985915492957744,
      "grad_norm": 1.8359375,
      "learning_rate": 5e-05,
      "loss": 1.2948389053344727,
      "mean_token_accuracy": 0.7675830125808716,
      "num_tokens": 125795.0,
      "step": 44
    },
    {
      "entropy": 1.3125,
      "epoch": 0.31690140845070425,
      "grad_norm": 1.875,
      "learning_rate": 5e-05,
      "loss": 1.2394049167633057,
      "mean_token_accuracy": 0.7697368264198303,
      "num_tokens": 128235.0,
      "step": 45
    },
    {
      "entropy": 1.3203125,
      "epoch": 0.323943661971831,
      "grad_norm": 1.6796875,
      "learning_rate": 5e-05,
      "loss": 1.3117918968200684,
      "mean_token_accuracy": 0.7587719559669495,
      "num_tokens": 131663.0,
      "step": 46
    },
    {
      "entropy": 1.3515625,
      "epoch": 0.33098591549295775,
      "grad_norm": 1.453125,
      "learning_rate": 5e-05,
      "loss": 1.3100031614303589,
      "mean_token_accuracy": 0.7717799544334412,
      "num_tokens": 134598.0,
      "step": 47
    },
    {
      "entropy": 1.3203125,
      "epoch": 0.3380281690140845,
      "grad_norm": 1.6953125,
      "learning_rate": 5e-05,
      "loss": 1.28191077709198,
      "mean_token_accuracy": 0.7734912037849426,
      "num_tokens": 137224.0,
      "step": 48
    },
    {
      "entropy": 1.3828125,
      "epoch": 0.34507042253521125,
      "grad_norm": 1.515625,
      "learning_rate": 5e-05,
      "loss": 1.3600674867630005,
      "mean_token_accuracy": 0.7438323497772217,
      "num_tokens": 140191.0,
      "step": 49
    },
    {
      "entropy": 1.21875,
      "epoch": 0.352112676056338,
      "grad_norm": 1.46875,
      "learning_rate": 5e-05,
      "loss": 1.1239631175994873,
      "mean_token_accuracy": 0.788614809513092,
      "num_tokens": 142834.0,
      "step": 50
    },
    {
      "entropy": 1.4765625,
      "epoch": 0.3591549295774648,
      "grad_norm": 1.46875,
      "learning_rate": 5e-05,
      "loss": 1.3836950063705444,
      "mean_token_accuracy": 0.7538405060768127,
      "num_tokens": 145576.0,
      "step": 51
    },
    {
      "entropy": 1.1328125,
      "epoch": 0.36619718309859156,
      "grad_norm": 1.59375,
      "learning_rate": 5e-05,
      "loss": 1.1020089387893677,
      "mean_token_accuracy": 0.8136998414993286,
      "num_tokens": 147876.0,
      "step": 52
    },
    {
      "entropy": 1.3671875,
      "epoch": 0.3732394366197183,
      "grad_norm": 1.15625,
      "learning_rate": 5e-05,
      "loss": 1.3127013444900513,
      "mean_token_accuracy": 0.7502166032791138,
      "num_tokens": 151347.0,
      "step": 53
    },
    {
      "entropy": 1.0390625,
      "epoch": 0.38028169014084506,
      "grad_norm": 1.25,
      "learning_rate": 5e-05,
      "loss": 1.0258498191833496,
      "mean_token_accuracy": 0.8231140971183777,
      "num_tokens": 154046.0,
      "step": 54
    },
    {
      "entropy": 1.296875,
      "epoch": 0.3873239436619718,
      "grad_norm": 1.515625,
      "learning_rate": 5e-05,
      "loss": 1.26340913772583,
      "mean_token_accuracy": 0.7880998253822327,
      "num_tokens": 156659.0,
      "step": 55
    },
    {
      "entropy": 1.3828125,
      "epoch": 0.39436619718309857,
      "grad_norm": 1.1484375,
      "learning_rate": 5e-05,
      "loss": 1.3492928743362427,
      "mean_token_accuracy": 0.7417179942131042,
      "num_tokens": 160078.0,
      "step": 56
    },
    {
      "entropy": 1.046875,
      "epoch": 0.4014084507042254,
      "grad_norm": 1.1171875,
      "learning_rate": 5e-05,
      "loss": 1.0268410444259644,
      "mean_token_accuracy": 0.8108739256858826,
      "num_tokens": 163084.0,
      "step": 57
    },
    {
      "entropy": 1.2734375,
      "epoch": 0.4084507042253521,
      "grad_norm": 1.296875,
      "learning_rate": 5e-05,
      "loss": 1.218086838722229,
      "mean_token_accuracy": 0.7687548398971558,
      "num_tokens": 165678.0,
      "step": 58
    },
    {
      "entropy": 1.265625,
      "epoch": 0.4154929577464789,
      "grad_norm": 1.078125,
      "learning_rate": 5e-05,
      "loss": 1.2634363174438477,
      "mean_token_accuracy": 0.7484593987464905,
      "num_tokens": 169256.0,
      "step": 59
    },
    {
      "entropy": 1.046875,
      "epoch": 0.4225352112676056,
      "grad_norm": 1.1796875,
      "learning_rate": 5e-05,
      "loss": 0.9637977480888367,
      "mean_token_accuracy": 0.8200221061706543,
      "num_tokens": 171981.0,
      "step": 60
    },
    {
      "entropy": 1.1015625,
      "epoch": 0.4295774647887324,
      "grad_norm": 1.09375,
      "learning_rate": 5e-05,
      "loss": 1.0381879806518555,
      "mean_token_accuracy": 0.8146411776542664,
      "num_tokens": 174762.0,
      "step": 61
    },
    {
      "entropy": 1.4609375,
      "epoch": 0.43661971830985913,
      "grad_norm": 1.109375,
      "learning_rate": 5e-05,
      "loss": 1.442441701889038,
      "mean_token_accuracy": 0.7236841917037964,
      "num_tokens": 177962.0,
      "step": 62
    },
    {
      "entropy": 0.9453125,
      "epoch": 0.44366197183098594,
      "grad_norm": 1.109375,
      "learning_rate": 5e-05,
      "loss": 0.9214452505111694,
      "mean_token_accuracy": 0.8193620443344116,
      "num_tokens": 180666.0,
      "step": 63
    },
    {
      "entropy": 1.0859375,
      "epoch": 0.4507042253521127,
      "grad_norm": 1.1484375,
      "learning_rate": 5e-05,
      "loss": 1.1097257137298584,
      "mean_token_accuracy": 0.8031823039054871,
      "num_tokens": 183565.0,
      "step": 64
    },
    {
      "entropy": 1.1640625,
      "epoch": 0.45774647887323944,
      "grad_norm": 1.140625,
      "learning_rate": 5e-05,
      "loss": 1.155249834060669,
      "mean_token_accuracy": 0.7887681126594543,
      "num_tokens": 186333.0,
      "step": 65
    },
    {
      "entropy": 0.85546875,
      "epoch": 0.4647887323943662,
      "grad_norm": 1.1953125,
      "learning_rate": 5e-05,
      "loss": 0.8293672800064087,
      "mean_token_accuracy": 0.8543021082878113,
      "num_tokens": 188956.0,
      "step": 66
    },
    {
      "entropy": 0.91015625,
      "epoch": 0.47183098591549294,
      "grad_norm": 1.1171875,
      "learning_rate": 5e-05,
      "loss": 0.894307553768158,
      "mean_token_accuracy": 0.8422175049781799,
      "num_tokens": 191309.0,
      "step": 67
    },
    {
      "entropy": 1.203125,
      "epoch": 0.4788732394366197,
      "grad_norm": 1.0234375,
      "learning_rate": 5e-05,
      "loss": 1.1631497144699097,
      "mean_token_accuracy": 0.7861995100975037,
      "num_tokens": 194114.0,
      "step": 68
    },
    {
      "entropy": 1.328125,
      "epoch": 0.4859154929577465,
      "grad_norm": 1.0390625,
      "learning_rate": 5e-05,
      "loss": 1.2809975147247314,
      "mean_token_accuracy": 0.7543198466300964,
      "num_tokens": 197305.0,
      "step": 69
    },
    {
      "entropy": 1.015625,
      "epoch": 0.49295774647887325,
      "grad_norm": 1.3203125,
      "learning_rate": 5e-05,
      "loss": 1.0155205726623535,
      "mean_token_accuracy": 0.834074079990387,
      "num_tokens": 199338.0,
      "step": 70
    },
    {
      "entropy": 0.83984375,
      "epoch": 0.5,
      "grad_norm": 1.1796875,
      "learning_rate": 5e-05,
      "loss": 0.7589200735092163,
      "mean_token_accuracy": 0.8618420958518982,
      "num_tokens": 201474.0,
      "step": 71
    },
    {
      "entropy": 1.0859375,
      "epoch": 0.5070422535211268,
      "grad_norm": 1.2109375,
      "learning_rate": 5e-05,
      "loss": 1.0090736150741577,
      "mean_token_accuracy": 0.8233413100242615,
      "num_tokens": 203984.0,
      "step": 72
    },
    {
      "entropy": 1.0703125,
      "epoch": 0.5140845070422535,
      "grad_norm": 0.9140625,
      "learning_rate": 5e-05,
      "loss": 1.046358346939087,
      "mean_token_accuracy": 0.802097737789154,
      "num_tokens": 207615.0,
      "step": 73
    },
    {
      "entropy": 1.203125,
      "epoch": 0.5211267605633803,
      "grad_norm": 1.2109375,
      "learning_rate": 5e-05,
      "loss": 1.14384126663208,
      "mean_token_accuracy": 0.7883082628250122,
      "num_tokens": 209881.0,
      "step": 74
    },
    {
      "entropy": 1.015625,
      "epoch": 0.528169014084507,
      "grad_norm": 0.98828125,
      "learning_rate": 5e-05,
      "loss": 0.9863561987876892,
      "mean_token_accuracy": 0.8165745735168457,
      "num_tokens": 212604.0,
      "step": 75
    },
    {
      "entropy": 1.2421875,
      "epoch": 0.5352112676056338,
      "grad_norm": 1.1171875,
      "learning_rate": 5e-05,
      "loss": 1.1881471872329712,
      "mean_token_accuracy": 0.7924298048019409,
      "num_tokens": 215069.0,
      "step": 76
    },
    {
      "entropy": 0.99609375,
      "epoch": 0.5422535211267606,
      "grad_norm": 0.86328125,
      "learning_rate": 5e-05,
      "loss": 0.9455610513687134,
      "mean_token_accuracy": 0.8238698244094849,
      "num_tokens": 217842.0,
      "step": 77
    },
    {
      "entropy": 0.93359375,
      "epoch": 0.5492957746478874,
      "grad_norm": 1.015625,
      "learning_rate": 5e-05,
      "loss": 0.8940377235412598,
      "mean_token_accuracy": 0.8325259685516357,
      "num_tokens": 220740.0,
      "step": 78
    },
    {
      "entropy": 1.078125,
      "epoch": 0.5563380281690141,
      "grad_norm": 1.0390625,
      "learning_rate": 5e-05,
      "loss": 1.072181224822998,
      "mean_token_accuracy": 0.7974197864532471,
      "num_tokens": 223616.0,
      "step": 79
    },
    {
      "entropy": 0.9609375,
      "epoch": 0.5633802816901409,
      "grad_norm": 1.0625,
      "learning_rate": 5e-05,
      "loss": 0.968547523021698,
      "mean_token_accuracy": 0.8254417181015015,
      "num_tokens": 226454.0,
      "step": 80
    },
    {
      "entropy": 1.0625,
      "epoch": 0.5704225352112676,
      "grad_norm": 1.0390625,
      "learning_rate": 5e-05,
      "loss": 1.029323935508728,
      "mean_token_accuracy": 0.8091970086097717,
      "num_tokens": 229376.0,
      "step": 81
    },
    {
      "entropy": 1.296875,
      "epoch": 0.5774647887323944,
      "grad_norm": 1.078125,
      "learning_rate": 5e-05,
      "loss": 1.30299711227417,
      "mean_token_accuracy": 0.7740401029586792,
      "num_tokens": 232327.0,
      "step": 82
    },
    {
      "entropy": 1.34375,
      "epoch": 0.5845070422535211,
      "grad_norm": 1.046875,
      "learning_rate": 5e-05,
      "loss": 1.3241432905197144,
      "mean_token_accuracy": 0.7471368312835693,
      "num_tokens": 235653.0,
      "step": 83
    },
    {
      "entropy": 1.1953125,
      "epoch": 0.5915492957746479,
      "grad_norm": 1.0546875,
      "learning_rate": 5e-05,
      "loss": 1.1841082572937012,
      "mean_token_accuracy": 0.7741320133209229,
      "num_tokens": 238282.0,
      "step": 84
    },
    {
      "entropy": 0.96875,
      "epoch": 0.5985915492957746,
      "grad_norm": 1.0,
      "learning_rate": 5e-05,
      "loss": 0.9051790833473206,
      "mean_token_accuracy": 0.8358089327812195,
      "num_tokens": 240781.0,
      "step": 85
    },
    {
      "entropy": 0.81640625,
      "epoch": 0.6056338028169014,
      "grad_norm": 1.0625,
      "learning_rate": 5e-05,
      "loss": 0.8058979511260986,
      "mean_token_accuracy": 0.8656787872314453,
      "num_tokens": 242881.0,
      "step": 86
    },
    {
      "entropy": 1.0234375,
      "epoch": 0.6126760563380281,
      "grad_norm": 0.89453125,
      "learning_rate": 5e-05,
      "loss": 0.9673476219177246,
      "mean_token_accuracy": 0.8199172616004944,
      "num_tokens": 246032.0,
      "step": 87
    },
    {
      "entropy": 0.9609375,
      "epoch": 0.6197183098591549,
      "grad_norm": 1.2265625,
      "learning_rate": 5e-05,
      "loss": 0.9231598377227783,
      "mean_token_accuracy": 0.833532452583313,
      "num_tokens": 248551.0,
      "step": 88
    },
    {
      "entropy": 0.9921875,
      "epoch": 0.6267605633802817,
      "grad_norm": 1.1015625,
      "learning_rate": 5e-05,
      "loss": 0.998257040977478,
      "mean_token_accuracy": 0.8266341686248779,
      "num_tokens": 251022.0,
      "step": 89
    },
    {
      "entropy": 1.0546875,
      "epoch": 0.6338028169014085,
      "grad_norm": 1.015625,
      "learning_rate": 5e-05,
      "loss": 1.0252176523208618,
      "mean_token_accuracy": 0.8216797709465027,
      "num_tokens": 253447.0,
      "step": 90
    },
    {
      "entropy": 1.1484375,
      "epoch": 0.6408450704225352,
      "grad_norm": 1.0,
      "learning_rate": 5e-05,
      "loss": 1.164228081703186,
      "mean_token_accuracy": 0.7702828645706177,
      "num_tokens": 256672.0,
      "step": 91
    },
    {
      "entropy": 0.77734375,
      "epoch": 0.647887323943662,
      "grad_norm": 0.9609375,
      "learning_rate": 5e-05,
      "loss": 0.7661694288253784,
      "mean_token_accuracy": 0.8448406457901001,
      "num_tokens": 259316.0,
      "step": 92
    },
    {
      "entropy": 1.15625,
      "epoch": 0.6549295774647887,
      "grad_norm": 0.98828125,
      "learning_rate": 5e-05,
      "loss": 1.1309537887573242,
      "mean_token_accuracy": 0.7973990440368652,
      "num_tokens": 262246.0,
      "step": 93
    },
    {
      "entropy": 1.046875,
      "epoch": 0.6619718309859155,
      "grad_norm": 0.9296875,
      "learning_rate": 5e-05,
      "loss": 1.0406370162963867,
      "mean_token_accuracy": 0.8099964261054993,
      "num_tokens": 265075.0,
      "step": 94
    },
    {
      "entropy": 1.0,
      "epoch": 0.6690140845070423,
      "grad_norm": 0.90234375,
      "learning_rate": 5e-05,
      "loss": 0.9478747844696045,
      "mean_token_accuracy": 0.8167044520378113,
      "num_tokens": 267729.0,
      "step": 95
    },
    {
      "entropy": 1.125,
      "epoch": 0.676056338028169,
      "grad_norm": 1.0234375,
      "learning_rate": 5e-05,
      "loss": 1.0806983709335327,
      "mean_token_accuracy": 0.8070116639137268,
      "num_tokens": 270732.0,
      "step": 96
    },
    {
      "entropy": 1.3359375,
      "epoch": 0.6830985915492958,
      "grad_norm": 0.95703125,
      "learning_rate": 5e-05,
      "loss": 1.2827727794647217,
      "mean_token_accuracy": 0.7681026458740234,
      "num_tokens": 274013.0,
      "step": 97
    },
    {
      "entropy": 0.96875,
      "epoch": 0.6901408450704225,
      "grad_norm": 1.1484375,
      "learning_rate": 5e-05,
      "loss": 0.9255345463752747,
      "mean_token_accuracy": 0.820858895778656,
      "num_tokens": 276466.0,
      "step": 98
    },
    {
      "entropy": 0.8828125,
      "epoch": 0.6971830985915493,
      "grad_norm": 0.828125,
      "learning_rate": 5e-05,
      "loss": 0.8368520736694336,
      "mean_token_accuracy": 0.8478841185569763,
      "num_tokens": 279097.0,
      "step": 99
    },
    {
      "entropy": 1.1875,
      "epoch": 0.704225352112676,
      "grad_norm": 0.91796875,
      "learning_rate": 5e-05,
      "loss": 1.1672570705413818,
      "mean_token_accuracy": 0.7689912915229797,
      "num_tokens": 282317.0,
      "step": 100
    },
    {
      "entropy": 1.53125,
      "epoch": 0.7112676056338029,
      "grad_norm": 1.0625,
      "learning_rate": 5e-05,
      "loss": 1.4877268075942993,
      "mean_token_accuracy": 0.7194092869758606,
      "num_tokens": 285643.0,
      "step": 101
    },
    {
      "entropy": 1.0,
      "epoch": 0.7183098591549296,
      "grad_norm": 1.0625,
      "learning_rate": 5e-05,
      "loss": 0.956112265586853,
      "mean_token_accuracy": 0.8256269693374634,
      "num_tokens": 288203.0,
      "step": 102
    },
    {
      "entropy": 1.1953125,
      "epoch": 0.7253521126760564,
      "grad_norm": 1.1328125,
      "learning_rate": 5e-05,
      "loss": 1.2514692544937134,
      "mean_token_accuracy": 0.771497905254364,
      "num_tokens": 291095.0,
      "step": 103
    },
    {
      "entropy": 1.2109375,
      "epoch": 0.7323943661971831,
      "grad_norm": 0.93359375,
      "learning_rate": 5e-05,
      "loss": 1.1472859382629395,
      "mean_token_accuracy": 0.7850604057312012,
      "num_tokens": 293834.0,
      "step": 104
    },
    {
      "entropy": 0.7734375,
      "epoch": 0.7394366197183099,
      "grad_norm": 1.28125,
      "learning_rate": 5e-05,
      "loss": 0.7254777550697327,
      "mean_token_accuracy": 0.8632119297981262,
      "num_tokens": 295984.0,
      "step": 105
    },
    {
      "entropy": 0.82421875,
      "epoch": 0.7464788732394366,
      "grad_norm": 1.078125,
      "learning_rate": 5e-05,
      "loss": 0.795278787612915,
      "mean_token_accuracy": 0.8470132946968079,
      "num_tokens": 298319.0,
      "step": 106
    },
    {
      "entropy": 1.09375,
      "epoch": 0.7535211267605634,
      "grad_norm": 0.87109375,
      "learning_rate": 5e-05,
      "loss": 1.0427790880203247,
      "mean_token_accuracy": 0.804815411567688,
      "num_tokens": 301442.0,
      "step": 107
    },
    {
      "entropy": 1.1171875,
      "epoch": 0.7605633802816901,
      "grad_norm": 0.95703125,
      "learning_rate": 5e-05,
      "loss": 1.104764699935913,
      "mean_token_accuracy": 0.8097593188285828,
      "num_tokens": 304483.0,
      "step": 108
    },
    {
      "entropy": 0.7421875,
      "epoch": 0.7676056338028169,
      "grad_norm": 1.0,
      "learning_rate": 5e-05,
      "loss": 0.6636262536048889,
      "mean_token_accuracy": 0.861122190952301,
      "num_tokens": 307004.0,
      "step": 109
    },
    {
      "entropy": 0.97265625,
      "epoch": 0.7746478873239436,
      "grad_norm": 0.90234375,
      "learning_rate": 5e-05,
      "loss": 0.9813203811645508,
      "mean_token_accuracy": 0.8114525079727173,
      "num_tokens": 309876.0,
      "step": 110
    },
    {
      "entropy": 1.09375,
      "epoch": 0.7816901408450704,
      "grad_norm": 1.078125,
      "learning_rate": 5e-05,
      "loss": 1.0708268880844116,
      "mean_token_accuracy": 0.7995486855506897,
      "num_tokens": 312543.0,
      "step": 111
    },
    {
      "entropy": 0.953125,
      "epoch": 0.7887323943661971,
      "grad_norm": 0.9453125,
      "learning_rate": 5e-05,
      "loss": 0.9298059344291687,
      "mean_token_accuracy": 0.8266475796699524,
      "num_tokens": 315343.0,
      "step": 112
    },
    {
      "entropy": 0.93359375,
      "epoch": 0.795774647887324,
      "grad_norm": 1.2109375,
      "learning_rate": 5e-05,
      "loss": 0.9146865606307983,
      "mean_token_accuracy": 0.8447837233543396,
      "num_tokens": 317709.0,
      "step": 113
    },
    {
      "entropy": 1.0390625,
      "epoch": 0.8028169014084507,
      "grad_norm": 1.015625,
      "learning_rate": 5e-05,
      "loss": 1.0023632049560547,
      "mean_token_accuracy": 0.8247303366661072,
      "num_tokens": 320313.0,
      "step": 114
    },
    {
      "entropy": 1.0234375,
      "epoch": 0.8098591549295775,
      "grad_norm": 0.9375,
      "learning_rate": 5e-05,
      "loss": 1.0119502544403076,
      "mean_token_accuracy": 0.8081619739532471,
      "num_tokens": 323433.0,
      "step": 115
    },
    {
      "entropy": 1.1484375,
      "epoch": 0.8169014084507042,
      "grad_norm": 1.09375,
      "learning_rate": 5e-05,
      "loss": 1.129537582397461,
      "mean_token_accuracy": 0.7915045619010925,
      "num_tokens": 326525.0,
      "step": 116
    },
    {
      "entropy": 0.828125,
      "epoch": 0.823943661971831,
      "grad_norm": 0.96875,
      "learning_rate": 5e-05,
      "loss": 0.8433165550231934,
      "mean_token_accuracy": 0.8409850001335144,
      "num_tokens": 328929.0,
      "step": 117
    },
    {
      "entropy": 1.0625,
      "epoch": 0.8309859154929577,
      "grad_norm": 1.0078125,
      "learning_rate": 5e-05,
      "loss": 1.0346664190292358,
      "mean_token_accuracy": 0.7833219170570374,
      "num_tokens": 331863.0,
      "step": 118
    },
    {
      "entropy": 0.98046875,
      "epoch": 0.8380281690140845,
      "grad_norm": 1.09375,
      "learning_rate": 5e-05,
      "loss": 0.950432300567627,
      "mean_token_accuracy": 0.8241989612579346,
      "num_tokens": 334243.0,
      "step": 119
    },
    {
      "entropy": 1.0546875,
      "epoch": 0.8450704225352113,
      "grad_norm": 0.8984375,
      "learning_rate": 5e-05,
      "loss": 1.0502727031707764,
      "mean_token_accuracy": 0.7917791604995728,
      "num_tokens": 337584.0,
      "step": 120
    },
    {
      "entropy": 0.89453125,
      "epoch": 0.852112676056338,
      "grad_norm": 0.9765625,
      "learning_rate": 5e-05,
      "loss": 0.9121181964874268,
      "mean_token_accuracy": 0.8309713006019592,
      "num_tokens": 340485.0,
      "step": 121
    },
    {
      "entropy": 0.88671875,
      "epoch": 0.8591549295774648,
      "grad_norm": 1.1328125,
      "learning_rate": 5e-05,
      "loss": 0.9219837188720703,
      "mean_token_accuracy": 0.8140540719032288,
      "num_tokens": 343268.0,
      "step": 122
    },
    {
      "entropy": 0.8515625,
      "epoch": 0.8661971830985915,
      "grad_norm": 1.1484375,
      "learning_rate": 5e-05,
      "loss": 0.7965174317359924,
      "mean_token_accuracy": 0.8607358932495117,
      "num_tokens": 345423.0,
      "step": 123
    },
    {
      "entropy": 0.9765625,
      "epoch": 0.8732394366197183,
      "grad_norm": 0.82421875,
      "learning_rate": 5e-05,
      "loss": 0.9578317999839783,
      "mean_token_accuracy": 0.8050633072853088,
      "num_tokens": 348591.0,
      "step": 124
    },
    {
      "entropy": 0.953125,
      "epoch": 0.8802816901408451,
      "grad_norm": 0.890625,
      "learning_rate": 5e-05,
      "loss": 0.9342446327209473,
      "mean_token_accuracy": 0.8261499404907227,
      "num_tokens": 351360.0,
      "step": 125
    },
    {
      "entropy": 1.0859375,
      "epoch": 0.8873239436619719,
      "grad_norm": 0.88671875,
      "learning_rate": 5e-05,
      "loss": 1.0417184829711914,
      "mean_token_accuracy": 0.7993496060371399,
      "num_tokens": 354443.0,
      "step": 126
    },
    {
      "entropy": 1.1875,
      "epoch": 0.8943661971830986,
      "grad_norm": 0.96484375,
      "learning_rate": 5e-05,
      "loss": 1.169641137123108,
      "mean_token_accuracy": 0.789621889591217,
      "num_tokens": 356937.0,
      "step": 127
    },
    {
      "entropy": 1.171875,
      "epoch": 0.9014084507042254,
      "grad_norm": 1.0546875,
      "learning_rate": 5e-05,
      "loss": 1.0926804542541504,
      "mean_token_accuracy": 0.78143709897995,
      "num_tokens": 359951.0,
      "step": 128
    },
    {
      "entropy": 0.90625,
      "epoch": 0.9084507042253521,
      "grad_norm": 1.046875,
      "learning_rate": 5e-05,
      "loss": 0.8831167817115784,
      "mean_token_accuracy": 0.8297297358512878,
      "num_tokens": 362549.0,
      "step": 129
    },
    {
      "entropy": 1.34375,
      "epoch": 0.9154929577464789,
      "grad_norm": 1.1640625,
      "learning_rate": 5e-05,
      "loss": 1.2982932329177856,
      "mean_token_accuracy": 0.7599736452102661,
      "num_tokens": 365590.0,
      "step": 130
    },
    {
      "entropy": 0.8515625,
      "epoch": 0.9225352112676056,
      "grad_norm": 1.140625,
      "learning_rate": 5e-05,
      "loss": 0.7926298975944519,
      "mean_token_accuracy": 0.8279292583465576,
      "num_tokens": 368312.0,
      "step": 131
    },
    {
      "entropy": 0.85546875,
      "epoch": 0.9295774647887324,
      "grad_norm": 1.0546875,
      "learning_rate": 5e-05,
      "loss": 0.876966655254364,
      "mean_token_accuracy": 0.8338558077812195,
      "num_tokens": 370872.0,
      "step": 132
    },
    {
      "entropy": 0.58203125,
      "epoch": 0.9366197183098591,
      "grad_norm": 1.09375,
      "learning_rate": 5e-05,
      "loss": 0.5232805609703064,
      "mean_token_accuracy": 0.9088762998580933,
      "num_tokens": 372998.0,
      "step": 133
    },
    {
      "entropy": 1.3125,
      "epoch": 0.9436619718309859,
      "grad_norm": 1.0859375,
      "learning_rate": 5e-05,
      "loss": 1.2858319282531738,
      "mean_token_accuracy": 0.764127790927887,
      "num_tokens": 375855.0,
      "step": 134
    },
    {
      "entropy": 0.9453125,
      "epoch": 0.9507042253521126,
      "grad_norm": 0.8125,
      "learning_rate": 5e-05,
      "loss": 0.9115528464317322,
      "mean_token_accuracy": 0.8201290965080261,
      "num_tokens": 379271.0,
      "step": 135
    },
    {
      "entropy": 1.0703125,
      "epoch": 0.9577464788732394,
      "grad_norm": 1.3828125,
      "learning_rate": 5e-05,
      "loss": 1.03813898563385,
      "mean_token_accuracy": 0.8119325637817383,
      "num_tokens": 381592.0,
      "step": 136
    },
    {
      "entropy": 1.109375,
      "epoch": 0.9647887323943662,
      "grad_norm": 0.953125,
      "learning_rate": 5e-05,
      "loss": 1.1146811246871948,
      "mean_token_accuracy": 0.790168285369873,
      "num_tokens": 384631.0,
      "step": 137
    },
    {
      "entropy": 1.1640625,
      "epoch": 0.971830985915493,
      "grad_norm": 1.0234375,
      "learning_rate": 5e-05,
      "loss": 1.173197627067566,
      "mean_token_accuracy": 0.7781087160110474,
      "num_tokens": 387325.0,
      "step": 138
    },
    {
      "entropy": 0.828125,
      "epoch": 0.9788732394366197,
      "grad_norm": 1.0546875,
      "learning_rate": 5e-05,
      "loss": 0.8895862698554993,
      "mean_token_accuracy": 0.8376736044883728,
      "num_tokens": 389637.0,
      "step": 139
    },
    {
      "entropy": 1.1953125,
      "epoch": 0.9859154929577465,
      "grad_norm": 0.8515625,
      "learning_rate": 5e-05,
      "loss": 1.2043834924697876,
      "mean_token_accuracy": 0.7506383061408997,
      "num_tokens": 393170.0,
      "step": 140
    },
    {
      "entropy": 0.87109375,
      "epoch": 0.9929577464788732,
      "grad_norm": 1.0546875,
      "learning_rate": 5e-05,
      "loss": 0.8573941588401794,
      "mean_token_accuracy": 0.8464516401290894,
      "num_tokens": 395503.0,
      "step": 141
    },
    {
      "entropy": 1.078125,
      "epoch": 1.0,
      "grad_norm": 0.90625,
      "learning_rate": 5e-05,
      "loss": 1.0620708465576172,
      "mean_token_accuracy": 0.7814034819602966,
      "num_tokens": 398361.0,
      "step": 142
    },
    {
      "epoch": 1.0,
      "eval_entropy": 0.9779188368055556,
      "eval_loss": 0.9645262956619263,
      "eval_mean_token_accuracy": 0.8152486599153943,
      "eval_num_tokens": 398361.0,
      "eval_runtime": 3.7052,
      "eval_samples_per_second": 76.65,
      "eval_steps_per_second": 9.716,
      "step": 142
    }
  ],
  "logging_steps": 1,
  "max_steps": 710,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 370240540913664.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}