{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008,
      "grad_norm": 37.127628326416016,
      "learning_rate": 1e-05,
      "loss": 15.0452,
      "mean_token_accuracy": 0.41070467978715897,
      "step": 1
    },
    {
      "epoch": 0.016,
      "grad_norm": 31.609861373901367,
      "learning_rate": 2e-05,
      "loss": 14.4012,
      "mean_token_accuracy": 0.4362717792391777,
      "step": 2
    },
    {
      "epoch": 0.024,
      "grad_norm": 28.683244705200195,
      "learning_rate": 3e-05,
      "loss": 13.963,
      "mean_token_accuracy": 0.4400983825325966,
      "step": 3
    },
    {
      "epoch": 0.032,
      "grad_norm": 22.16831398010254,
      "learning_rate": 4e-05,
      "loss": 13.3495,
      "mean_token_accuracy": 0.4725849777460098,
      "step": 4
    },
    {
      "epoch": 0.04,
      "grad_norm": 19.798992156982422,
      "learning_rate": 5e-05,
      "loss": 11.6453,
      "mean_token_accuracy": 0.5295183658599854,
      "step": 5
    },
    {
      "epoch": 0.048,
      "grad_norm": 17.516647338867188,
      "learning_rate": 4.98989898989899e-05,
      "loss": 11.3724,
      "mean_token_accuracy": 0.5440531671047211,
      "step": 6
    },
    {
      "epoch": 0.056,
      "grad_norm": 18.01726531982422,
      "learning_rate": 4.97979797979798e-05,
      "loss": 10.463,
      "mean_token_accuracy": 0.5659109503030777,
      "step": 7
    },
    {
      "epoch": 0.064,
      "grad_norm": 16.347856521606445,
      "learning_rate": 4.9696969696969694e-05,
      "loss": 9.83,
      "mean_token_accuracy": 0.5818765759468079,
      "step": 8
    },
    {
      "epoch": 0.072,
      "grad_norm": 17.102680206298828,
      "learning_rate": 4.9595959595959594e-05,
      "loss": 9.21,
      "mean_token_accuracy": 0.6048124134540558,
      "step": 9
    },
    {
      "epoch": 0.08,
      "grad_norm": 13.707213401794434,
      "learning_rate": 4.94949494949495e-05,
      "loss": 9.0634,
      "mean_token_accuracy": 0.6290689557790756,
      "step": 10
    },
    {
      "epoch": 0.088,
      "grad_norm": 12.317888259887695,
      "learning_rate": 4.93939393939394e-05,
      "loss": 8.7963,
      "mean_token_accuracy": 0.6293386816978455,
      "step": 11
    },
    {
      "epoch": 0.096,
      "grad_norm": 12.71199893951416,
      "learning_rate": 4.92929292929293e-05,
      "loss": 7.9924,
      "mean_token_accuracy": 0.6720790416002274,
      "step": 12
    },
    {
      "epoch": 0.104,
      "grad_norm": 11.422212600708008,
      "learning_rate": 4.919191919191919e-05,
      "loss": 8.0788,
      "mean_token_accuracy": 0.6582284867763519,
      "step": 13
    },
    {
      "epoch": 0.112,
      "grad_norm": 10.79257869720459,
      "learning_rate": 4.909090909090909e-05,
      "loss": 7.5785,
      "mean_token_accuracy": 0.673900306224823,
      "step": 14
    },
    {
      "epoch": 0.12,
      "grad_norm": 10.639641761779785,
      "learning_rate": 4.898989898989899e-05,
      "loss": 7.3491,
      "mean_token_accuracy": 0.6823591589927673,
      "step": 15
    },
    {
      "epoch": 0.128,
      "grad_norm": 11.140637397766113,
      "learning_rate": 4.888888888888889e-05,
      "loss": 7.6078,
      "mean_token_accuracy": 0.6652624905109406,
      "step": 16
    },
    {
      "epoch": 0.136,
      "grad_norm": 10.391518592834473,
      "learning_rate": 4.878787878787879e-05,
      "loss": 7.1951,
      "mean_token_accuracy": 0.66854228079319,
      "step": 17
    },
    {
      "epoch": 0.144,
      "grad_norm": 10.393790245056152,
      "learning_rate": 4.868686868686869e-05,
      "loss": 7.1549,
      "mean_token_accuracy": 0.6921204030513763,
      "step": 18
    },
    {
      "epoch": 0.152,
      "grad_norm": 9.204154014587402,
      "learning_rate": 4.858585858585859e-05,
      "loss": 7.3606,
      "mean_token_accuracy": 0.6926577091217041,
      "step": 19
    },
    {
      "epoch": 0.16,
      "grad_norm": 10.39026927947998,
      "learning_rate": 4.848484848484849e-05,
      "loss": 7.0433,
      "mean_token_accuracy": 0.6972606927156448,
      "step": 20
    },
    {
      "epoch": 0.168,
      "grad_norm": 10.690828323364258,
      "learning_rate": 4.838383838383839e-05,
      "loss": 7.0197,
      "mean_token_accuracy": 0.6896309554576874,
      "step": 21
    },
    {
      "epoch": 0.176,
      "grad_norm": 10.476662635803223,
      "learning_rate": 4.828282828282829e-05,
      "loss": 6.2416,
      "mean_token_accuracy": 0.7330080419778824,
      "step": 22
    },
    {
      "epoch": 0.184,
      "grad_norm": 9.495307922363281,
      "learning_rate": 4.8181818181818186e-05,
      "loss": 6.7721,
      "mean_token_accuracy": 0.6969246119260788,
      "step": 23
    },
    {
      "epoch": 0.192,
      "grad_norm": 9.304316520690918,
      "learning_rate": 4.808080808080808e-05,
      "loss": 6.4576,
      "mean_token_accuracy": 0.7197146117687225,
      "step": 24
    },
    {
      "epoch": 0.2,
      "grad_norm": 9.24569034576416,
      "learning_rate": 4.797979797979798e-05,
      "loss": 5.9051,
      "mean_token_accuracy": 0.7319369614124298,
      "step": 25
    },
    {
      "epoch": 0.208,
      "grad_norm": 8.99447250366211,
      "learning_rate": 4.787878787878788e-05,
      "loss": 6.113,
      "mean_token_accuracy": 0.7273330986499786,
      "step": 26
    },
    {
      "epoch": 0.216,
      "grad_norm": 9.245104789733887,
      "learning_rate": 4.7777777777777784e-05,
      "loss": 6.5231,
      "mean_token_accuracy": 0.7232353389263153,
      "step": 27
    },
    {
      "epoch": 0.224,
      "grad_norm": 9.193451881408691,
      "learning_rate": 4.7676767676767684e-05,
      "loss": 6.0893,
      "mean_token_accuracy": 0.7589968591928482,
      "step": 28
    },
    {
      "epoch": 0.232,
      "grad_norm": 9.738916397094727,
      "learning_rate": 4.7575757575757576e-05,
      "loss": 6.2512,
      "mean_token_accuracy": 0.7394937723875046,
      "step": 29
    },
    {
      "epoch": 0.24,
      "grad_norm": 9.301070213317871,
      "learning_rate": 4.7474747474747476e-05,
      "loss": 6.4148,
      "mean_token_accuracy": 0.7273970693349838,
      "step": 30
    },
    {
      "epoch": 0.248,
      "grad_norm": 9.732291221618652,
      "learning_rate": 4.7373737373737375e-05,
      "loss": 6.4581,
      "mean_token_accuracy": 0.7421613037586212,
      "step": 31
    },
    {
      "epoch": 0.256,
      "grad_norm": 9.543816566467285,
      "learning_rate": 4.7272727272727275e-05,
      "loss": 5.7812,
      "mean_token_accuracy": 0.7518225610256195,
      "step": 32
    },
    {
      "epoch": 0.264,
      "grad_norm": 8.442834854125977,
      "learning_rate": 4.7171717171717174e-05,
      "loss": 5.7968,
      "mean_token_accuracy": 0.7630135715007782,
      "step": 33
    },
    {
      "epoch": 0.272,
      "grad_norm": 7.911731243133545,
      "learning_rate": 4.7070707070707074e-05,
      "loss": 5.9594,
      "mean_token_accuracy": 0.7520367801189423,
      "step": 34
    },
    {
      "epoch": 0.28,
      "grad_norm": 8.111591339111328,
      "learning_rate": 4.696969696969697e-05,
      "loss": 6.1152,
      "mean_token_accuracy": 0.7495731711387634,
      "step": 35
    },
    {
      "epoch": 0.288,
      "grad_norm": 7.632225036621094,
      "learning_rate": 4.686868686868687e-05,
      "loss": 5.978,
      "mean_token_accuracy": 0.7462944090366364,
      "step": 36
    },
    {
      "epoch": 0.296,
      "grad_norm": 7.140631675720215,
      "learning_rate": 4.676767676767677e-05,
      "loss": 5.4284,
      "mean_token_accuracy": 0.7744311541318893,
      "step": 37
    },
    {
      "epoch": 0.304,
      "grad_norm": 7.504866600036621,
      "learning_rate": 4.666666666666667e-05,
      "loss": 5.4771,
      "mean_token_accuracy": 0.7578010857105255,
      "step": 38
    },
    {
      "epoch": 0.312,
      "grad_norm": 6.958098411560059,
      "learning_rate": 4.656565656565657e-05,
      "loss": 6.0761,
      "mean_token_accuracy": 0.7286622673273087,
      "step": 39
    },
    {
      "epoch": 0.32,
      "grad_norm": 7.4256696701049805,
      "learning_rate": 4.6464646464646464e-05,
      "loss": 5.9964,
      "mean_token_accuracy": 0.7376932203769684,
      "step": 40
    },
    {
      "epoch": 0.328,
      "grad_norm": 6.8756608963012695,
      "learning_rate": 4.636363636363636e-05,
      "loss": 5.2819,
      "mean_token_accuracy": 0.7755442261695862,
      "step": 41
    },
    {
      "epoch": 0.336,
      "grad_norm": 6.877252101898193,
      "learning_rate": 4.626262626262626e-05,
      "loss": 5.6559,
      "mean_token_accuracy": 0.7578635513782501,
      "step": 42
    },
    {
      "epoch": 0.344,
      "grad_norm": 7.236316680908203,
      "learning_rate": 4.616161616161616e-05,
      "loss": 5.3482,
      "mean_token_accuracy": 0.758240357041359,
      "step": 43
    },
    {
      "epoch": 0.352,
      "grad_norm": 7.695366859436035,
      "learning_rate": 4.606060606060607e-05,
      "loss": 5.3286,
      "mean_token_accuracy": 0.769702136516571,
      "step": 44
    },
    {
      "epoch": 0.36,
      "grad_norm": 7.2449164390563965,
      "learning_rate": 4.595959595959596e-05,
      "loss": 5.3446,
      "mean_token_accuracy": 0.750788614153862,
      "step": 45
    },
    {
      "epoch": 0.368,
      "grad_norm": 6.902708530426025,
      "learning_rate": 4.585858585858586e-05,
      "loss": 4.5337,
      "mean_token_accuracy": 0.7888868898153305,
      "step": 46
    },
    {
      "epoch": 0.376,
      "grad_norm": 7.465998649597168,
      "learning_rate": 4.575757575757576e-05,
      "loss": 5.2185,
      "mean_token_accuracy": 0.7631554305553436,
      "step": 47
    },
    {
      "epoch": 0.384,
      "grad_norm": 6.581245422363281,
      "learning_rate": 4.565656565656566e-05,
      "loss": 5.6984,
      "mean_token_accuracy": 0.7439078986644745,
      "step": 48
    },
    {
      "epoch": 0.392,
      "grad_norm": 6.623785495758057,
      "learning_rate": 4.555555555555556e-05,
      "loss": 5.5339,
      "mean_token_accuracy": 0.7492197006940842,
      "step": 49
    },
    {
      "epoch": 0.4,
      "grad_norm": 7.161900997161865,
      "learning_rate": 4.545454545454546e-05,
      "loss": 6.0623,
      "mean_token_accuracy": 0.7385092377662659,
      "step": 50
    },
    {
      "epoch": 0.408,
      "grad_norm": 7.114354133605957,
      "learning_rate": 4.535353535353535e-05,
      "loss": 5.0984,
      "mean_token_accuracy": 0.7639837712049484,
      "step": 51
    },
    {
      "epoch": 0.416,
      "grad_norm": 6.865662097930908,
      "learning_rate": 4.525252525252526e-05,
      "loss": 4.5515,
      "mean_token_accuracy": 0.7843715995550156,
      "step": 52
    },
    {
      "epoch": 0.424,
      "grad_norm": 6.6706109046936035,
      "learning_rate": 4.515151515151516e-05,
      "loss": 5.7281,
      "mean_token_accuracy": 0.7444053590297699,
      "step": 53
    },
    {
      "epoch": 0.432,
      "grad_norm": 7.425290107727051,
      "learning_rate": 4.5050505050505056e-05,
      "loss": 5.5055,
      "mean_token_accuracy": 0.7495678812265396,
      "step": 54
    },
    {
      "epoch": 0.44,
      "grad_norm": 7.8452277183532715,
      "learning_rate": 4.494949494949495e-05,
      "loss": 5.5757,
      "mean_token_accuracy": 0.7629143446683884,
      "step": 55
    },
    {
      "epoch": 0.448,
      "grad_norm": 7.112468242645264,
      "learning_rate": 4.484848484848485e-05,
      "loss": 4.6575,
      "mean_token_accuracy": 0.7812573164701462,
      "step": 56
    },
    {
      "epoch": 0.456,
      "grad_norm": 6.542463302612305,
      "learning_rate": 4.474747474747475e-05,
      "loss": 5.373,
      "mean_token_accuracy": 0.7609668523073196,
      "step": 57
    },
    {
      "epoch": 0.464,
      "grad_norm": 6.580011367797852,
      "learning_rate": 4.464646464646465e-05,
      "loss": 4.5268,
      "mean_token_accuracy": 0.7902613431215286,
      "step": 58
    },
    {
      "epoch": 0.472,
      "grad_norm": 6.447749614715576,
      "learning_rate": 4.454545454545455e-05,
      "loss": 4.948,
      "mean_token_accuracy": 0.7766893953084946,
      "step": 59
    },
    {
      "epoch": 0.48,
      "grad_norm": 6.1748199462890625,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 4.8793,
      "mean_token_accuracy": 0.78531713783741,
      "step": 60
    },
    {
      "epoch": 0.488,
      "grad_norm": 7.2349114418029785,
      "learning_rate": 4.4343434343434346e-05,
      "loss": 5.4005,
      "mean_token_accuracy": 0.7713855803012848,
      "step": 61
    },
    {
      "epoch": 0.496,
      "grad_norm": 7.137668132781982,
      "learning_rate": 4.4242424242424246e-05,
      "loss": 4.7612,
      "mean_token_accuracy": 0.7853883504867554,
      "step": 62
    },
    {
      "epoch": 0.504,
      "grad_norm": 6.94537353515625,
      "learning_rate": 4.4141414141414145e-05,
      "loss": 5.0952,
      "mean_token_accuracy": 0.7686220556497574,
      "step": 63
    },
    {
      "epoch": 0.512,
      "grad_norm": 7.088626861572266,
      "learning_rate": 4.4040404040404044e-05,
      "loss": 5.107,
      "mean_token_accuracy": 0.7601823508739471,
      "step": 64
    },
    {
      "epoch": 0.52,
      "grad_norm": 6.948323726654053,
      "learning_rate": 4.3939393939393944e-05,
      "loss": 4.9925,
      "mean_token_accuracy": 0.7808158993721008,
      "step": 65
    },
    {
      "epoch": 0.528,
      "grad_norm": 6.151853084564209,
      "learning_rate": 4.383838383838384e-05,
      "loss": 4.7631,
      "mean_token_accuracy": 0.7758619040250778,
      "step": 66
    },
    {
      "epoch": 0.536,
      "grad_norm": 6.796617031097412,
      "learning_rate": 4.3737373737373736e-05,
      "loss": 5.1669,
      "mean_token_accuracy": 0.7829089462757111,
      "step": 67
    },
    {
      "epoch": 0.544,
      "grad_norm": 6.727445602416992,
      "learning_rate": 4.3636363636363636e-05,
      "loss": 5.9112,
      "mean_token_accuracy": 0.7398245632648468,
      "step": 68
    },
    {
      "epoch": 0.552,
      "grad_norm": 6.673349380493164,
      "learning_rate": 4.3535353535353535e-05,
      "loss": 5.5631,
      "mean_token_accuracy": 0.7459984719753265,
      "step": 69
    },
    {
      "epoch": 0.56,
      "grad_norm": 6.960887432098389,
      "learning_rate": 4.343434343434344e-05,
      "loss": 4.8842,
      "mean_token_accuracy": 0.7681605517864227,
      "step": 70
    },
    {
      "epoch": 0.568,
      "grad_norm": 6.884949684143066,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 4.9392,
      "mean_token_accuracy": 0.7818175554275513,
      "step": 71
    },
    {
      "epoch": 0.576,
      "grad_norm": 6.305865287780762,
      "learning_rate": 4.3232323232323234e-05,
      "loss": 4.3928,
      "mean_token_accuracy": 0.7899681925773621,
      "step": 72
    },
    {
      "epoch": 0.584,
      "grad_norm": 6.652983665466309,
      "learning_rate": 4.313131313131313e-05,
      "loss": 5.6515,
      "mean_token_accuracy": 0.7437300831079483,
      "step": 73
    },
    {
      "epoch": 0.592,
      "grad_norm": 5.960973262786865,
      "learning_rate": 4.303030303030303e-05,
      "loss": 4.2172,
      "mean_token_accuracy": 0.8126555383205414,
      "step": 74
    },
    {
      "epoch": 0.6,
      "grad_norm": 6.401238918304443,
      "learning_rate": 4.292929292929293e-05,
      "loss": 4.6772,
      "mean_token_accuracy": 0.7884032130241394,
      "step": 75
    },
    {
      "epoch": 0.608,
      "grad_norm": 6.02972412109375,
      "learning_rate": 4.282828282828283e-05,
      "loss": 4.7759,
      "mean_token_accuracy": 0.7719396352767944,
      "step": 76
    },
    {
      "epoch": 0.616,
      "grad_norm": 5.91736364364624,
      "learning_rate": 4.2727272727272724e-05,
      "loss": 4.5561,
      "mean_token_accuracy": 0.8052177727222443,
      "step": 77
    },
    {
      "epoch": 0.624,
      "grad_norm": 6.276677131652832,
      "learning_rate": 4.262626262626263e-05,
      "loss": 4.5458,
      "mean_token_accuracy": 0.7864794135093689,
      "step": 78
    },
    {
      "epoch": 0.632,
      "grad_norm": 6.683462142944336,
      "learning_rate": 4.252525252525253e-05,
      "loss": 4.7156,
      "mean_token_accuracy": 0.7799786478281021,
      "step": 79
    },
    {
      "epoch": 0.64,
      "grad_norm": 6.73935079574585,
      "learning_rate": 4.242424242424243e-05,
      "loss": 4.6576,
      "mean_token_accuracy": 0.7870494276285172,
      "step": 80
    },
    {
      "epoch": 0.648,
      "grad_norm": 6.009264945983887,
      "learning_rate": 4.232323232323233e-05,
      "loss": 4.3781,
      "mean_token_accuracy": 0.8029303699731827,
      "step": 81
    },
    {
      "epoch": 0.656,
      "grad_norm": 5.8821797370910645,
      "learning_rate": 4.222222222222222e-05,
      "loss": 4.6497,
      "mean_token_accuracy": 0.7830232828855515,
      "step": 82
    },
    {
      "epoch": 0.664,
      "grad_norm": 6.2794904708862305,
      "learning_rate": 4.212121212121212e-05,
      "loss": 4.563,
      "mean_token_accuracy": 0.7883824855089188,
      "step": 83
    },
    {
      "epoch": 0.672,
      "grad_norm": 6.4131364822387695,
      "learning_rate": 4.202020202020202e-05,
      "loss": 4.86,
      "mean_token_accuracy": 0.7777279317378998,
      "step": 84
    },
    {
      "epoch": 0.68,
      "grad_norm": 6.024682998657227,
      "learning_rate": 4.191919191919192e-05,
      "loss": 4.5301,
      "mean_token_accuracy": 0.8029981702566147,
      "step": 85
    },
    {
      "epoch": 0.688,
      "grad_norm": 6.400689125061035,
      "learning_rate": 4.181818181818182e-05,
      "loss": 4.5095,
      "mean_token_accuracy": 0.7834673821926117,
      "step": 86
    },
    {
      "epoch": 0.696,
      "grad_norm": 6.309573173522949,
      "learning_rate": 4.171717171717172e-05,
      "loss": 4.9511,
      "mean_token_accuracy": 0.7793049812316895,
      "step": 87
    },
    {
      "epoch": 0.704,
      "grad_norm": 6.098147869110107,
      "learning_rate": 4.161616161616162e-05,
      "loss": 4.8193,
      "mean_token_accuracy": 0.7891133576631546,
      "step": 88
    },
    {
      "epoch": 0.712,
      "grad_norm": 6.305475234985352,
      "learning_rate": 4.151515151515152e-05,
      "loss": 5.2867,
      "mean_token_accuracy": 0.763544350862503,
      "step": 89
    },
    {
      "epoch": 0.72,
      "grad_norm": 5.927306175231934,
      "learning_rate": 4.141414141414142e-05,
      "loss": 4.6925,
      "mean_token_accuracy": 0.7907081097364426,
      "step": 90
    },
    {
      "epoch": 0.728,
      "grad_norm": 6.21611213684082,
      "learning_rate": 4.131313131313132e-05,
      "loss": 4.1949,
      "mean_token_accuracy": 0.7960730195045471,
      "step": 91
    },
    {
      "epoch": 0.736,
      "grad_norm": 5.893104076385498,
      "learning_rate": 4.1212121212121216e-05,
      "loss": 4.3789,
      "mean_token_accuracy": 0.8057213127613068,
      "step": 92
    },
    {
      "epoch": 0.744,
      "grad_norm": 6.116573333740234,
      "learning_rate": 4.111111111111111e-05,
      "loss": 4.7077,
      "mean_token_accuracy": 0.7898276895284653,
      "step": 93
    },
    {
      "epoch": 0.752,
      "grad_norm": 5.917790412902832,
      "learning_rate": 4.101010101010101e-05,
      "loss": 4.3604,
      "mean_token_accuracy": 0.7916617542505264,
      "step": 94
    },
    {
      "epoch": 0.76,
      "grad_norm": 6.027205944061279,
      "learning_rate": 4.0909090909090915e-05,
      "loss": 4.3654,
      "mean_token_accuracy": 0.7899818271398544,
      "step": 95
    },
    {
      "epoch": 0.768,
      "grad_norm": 6.685177326202393,
      "learning_rate": 4.0808080808080814e-05,
      "loss": 4.3956,
      "mean_token_accuracy": 0.7992375791072845,
      "step": 96
    },
    {
      "epoch": 0.776,
      "grad_norm": 6.8115081787109375,
      "learning_rate": 4.070707070707071e-05,
      "loss": 4.4709,
      "mean_token_accuracy": 0.7787137180566788,
      "step": 97
    },
    {
      "epoch": 0.784,
      "grad_norm": 6.106410980224609,
      "learning_rate": 4.0606060606060606e-05,
      "loss": 4.2777,
      "mean_token_accuracy": 0.8080956637859344,
      "step": 98
    },
    {
      "epoch": 0.792,
      "grad_norm": 6.435486793518066,
      "learning_rate": 4.0505050505050506e-05,
      "loss": 4.7062,
      "mean_token_accuracy": 0.7773082256317139,
      "step": 99
    },
    {
      "epoch": 0.8,
      "grad_norm": 6.885725498199463,
      "learning_rate": 4.0404040404040405e-05,
      "loss": 4.7163,
      "mean_token_accuracy": 0.7818236798048019,
      "step": 100
    },
    {
      "epoch": 0.808,
      "grad_norm": 5.755829811096191,
      "learning_rate": 4.0303030303030305e-05,
      "loss": 4.4307,
      "mean_token_accuracy": 0.7950419485569,
      "step": 101
    },
    {
      "epoch": 0.816,
      "grad_norm": 6.709305763244629,
      "learning_rate": 4.0202020202020204e-05,
      "loss": 4.2539,
      "mean_token_accuracy": 0.8147921562194824,
      "step": 102
    },
    {
      "epoch": 0.824,
      "grad_norm": 6.947343826293945,
      "learning_rate": 4.01010101010101e-05,
      "loss": 4.3526,
      "mean_token_accuracy": 0.7921736389398575,
      "step": 103
    },
    {
      "epoch": 0.832,
      "grad_norm": 6.356295108795166,
      "learning_rate": 4e-05,
      "loss": 4.8007,
      "mean_token_accuracy": 0.780563622713089,
      "step": 104
    },
    {
      "epoch": 0.84,
      "grad_norm": 5.941018104553223,
      "learning_rate": 3.98989898989899e-05,
      "loss": 4.3688,
      "mean_token_accuracy": 0.7866577059030533,
      "step": 105
    },
    {
      "epoch": 0.848,
      "grad_norm": 6.366528034210205,
      "learning_rate": 3.97979797979798e-05,
      "loss": 4.252,
      "mean_token_accuracy": 0.799225702881813,
      "step": 106
    },
    {
      "epoch": 0.856,
      "grad_norm": 6.697314739227295,
      "learning_rate": 3.96969696969697e-05,
      "loss": 4.8091,
      "mean_token_accuracy": 0.7797738015651703,
      "step": 107
    },
    {
      "epoch": 0.864,
      "grad_norm": 6.910213947296143,
      "learning_rate": 3.9595959595959594e-05,
      "loss": 4.8069,
      "mean_token_accuracy": 0.7732720673084259,
      "step": 108
    },
    {
      "epoch": 0.872,
      "grad_norm": 5.771172523498535,
      "learning_rate": 3.9494949494949494e-05,
      "loss": 4.2787,
      "mean_token_accuracy": 0.7975706905126572,
      "step": 109
    },
    {
      "epoch": 0.88,
      "grad_norm": 5.44094181060791,
      "learning_rate": 3.939393939393939e-05,
      "loss": 4.0384,
      "mean_token_accuracy": 0.8148421049118042,
      "step": 110
    },
    {
      "epoch": 0.888,
      "grad_norm": 6.703949451446533,
      "learning_rate": 3.929292929292929e-05,
      "loss": 4.7667,
      "mean_token_accuracy": 0.7871679961681366,
      "step": 111
    },
    {
      "epoch": 0.896,
      "grad_norm": 6.386756896972656,
      "learning_rate": 3.91919191919192e-05,
      "loss": 4.2787,
      "mean_token_accuracy": 0.7989258021116257,
      "step": 112
    },
    {
      "epoch": 0.904,
      "grad_norm": 5.399852275848389,
      "learning_rate": 3.909090909090909e-05,
      "loss": 3.9899,
      "mean_token_accuracy": 0.8192960321903229,
      "step": 113
    },
    {
      "epoch": 0.912,
      "grad_norm": 5.751123905181885,
      "learning_rate": 3.898989898989899e-05,
      "loss": 4.292,
      "mean_token_accuracy": 0.8131812363862991,
      "step": 114
    },
    {
      "epoch": 0.92,
      "grad_norm": 5.929274559020996,
      "learning_rate": 3.888888888888889e-05,
      "loss": 4.5909,
      "mean_token_accuracy": 0.7861783355474472,
      "step": 115
    },
    {
      "epoch": 0.928,
      "grad_norm": 6.452223777770996,
      "learning_rate": 3.878787878787879e-05,
      "loss": 4.3544,
      "mean_token_accuracy": 0.7967888861894608,
      "step": 116
    },
    {
      "epoch": 0.936,
      "grad_norm": 5.868208408355713,
      "learning_rate": 3.868686868686869e-05,
      "loss": 4.9455,
      "mean_token_accuracy": 0.7758602350950241,
      "step": 117
    },
    {
      "epoch": 0.944,
      "grad_norm": 6.712337493896484,
      "learning_rate": 3.858585858585859e-05,
      "loss": 4.275,
      "mean_token_accuracy": 0.8086762726306915,
      "step": 118
    },
    {
      "epoch": 0.952,
      "grad_norm": 5.891403675079346,
      "learning_rate": 3.848484848484848e-05,
      "loss": 3.9505,
      "mean_token_accuracy": 0.8316792696714401,
      "step": 119
    },
    {
      "epoch": 0.96,
      "grad_norm": 6.26602840423584,
      "learning_rate": 3.838383838383838e-05,
      "loss": 4.4454,
      "mean_token_accuracy": 0.7933251559734344,
      "step": 120
    },
    {
      "epoch": 0.968,
      "grad_norm": 5.930867671966553,
      "learning_rate": 3.828282828282829e-05,
      "loss": 4.346,
      "mean_token_accuracy": 0.7956322878599167,
      "step": 121
    },
    {
      "epoch": 0.976,
      "grad_norm": 6.317086219787598,
      "learning_rate": 3.818181818181819e-05,
      "loss": 3.9196,
      "mean_token_accuracy": 0.8177185207605362,
      "step": 122
    },
    {
      "epoch": 0.984,
      "grad_norm": 6.2951178550720215,
      "learning_rate": 3.8080808080808087e-05,
      "loss": 4.3412,
      "mean_token_accuracy": 0.8002329915761948,
      "step": 123
    },
    {
      "epoch": 0.992,
      "grad_norm": 5.812499523162842,
      "learning_rate": 3.797979797979798e-05,
      "loss": 4.2135,
      "mean_token_accuracy": 0.8055464327335358,
      "step": 124
    },
    {
      "epoch": 1.0,
      "grad_norm": 6.186211109161377,
      "learning_rate": 3.787878787878788e-05,
      "loss": 4.2378,
      "mean_token_accuracy": 0.7995122522115707,
      "step": 125
    },
    {
      "epoch": 1.008,
      "grad_norm": 5.81443452835083,
      "learning_rate": 3.777777777777778e-05,
      "loss": 4.1608,
      "mean_token_accuracy": 0.80469611287117,
      "step": 126
    },
    {
      "epoch": 1.016,
      "grad_norm": 5.983344078063965,
      "learning_rate": 3.767676767676768e-05,
      "loss": 3.996,
      "mean_token_accuracy": 0.7908979803323746,
      "step": 127
    },
    {
      "epoch": 1.024,
      "grad_norm": 5.305781841278076,
      "learning_rate": 3.757575757575758e-05,
      "loss": 4.4283,
      "mean_token_accuracy": 0.7888950854539871,
      "step": 128
    },
    {
      "epoch": 1.032,
      "grad_norm": 5.880563259124756,
      "learning_rate": 3.747474747474748e-05,
      "loss": 4.2988,
      "mean_token_accuracy": 0.7900789678096771,
      "step": 129
    },
    {
      "epoch": 1.04,
      "grad_norm": 5.588067054748535,
      "learning_rate": 3.7373737373737376e-05,
      "loss": 3.5971,
      "mean_token_accuracy": 0.829292356967926,
      "step": 130
    },
    {
      "epoch": 1.048,
      "grad_norm": 5.570093631744385,
      "learning_rate": 3.7272727272727276e-05,
      "loss": 3.9778,
      "mean_token_accuracy": 0.814909428358078,
      "step": 131
    },
    {
      "epoch": 1.056,
      "grad_norm": 5.66859245300293,
      "learning_rate": 3.7171717171717175e-05,
      "loss": 3.8962,
      "mean_token_accuracy": 0.8082331418991089,
      "step": 132
    },
    {
      "epoch": 1.064,
      "grad_norm": 5.667624473571777,
      "learning_rate": 3.7070707070707075e-05,
      "loss": 3.3282,
      "mean_token_accuracy": 0.8463505655527115,
      "step": 133
    },
    {
      "epoch": 1.072,
      "grad_norm": 5.888784408569336,
      "learning_rate": 3.6969696969696974e-05,
      "loss": 4.236,
      "mean_token_accuracy": 0.7955542504787445,
      "step": 134
    },
    {
      "epoch": 1.08,
      "grad_norm": 6.280521392822266,
      "learning_rate": 3.686868686868687e-05,
      "loss": 3.8402,
      "mean_token_accuracy": 0.8112962692975998,
      "step": 135
    },
    {
      "epoch": 1.088,
      "grad_norm": 5.922068119049072,
      "learning_rate": 3.6767676767676766e-05,
      "loss": 3.8136,
      "mean_token_accuracy": 0.8134618252515793,
      "step": 136
    },
    {
      "epoch": 1.096,
      "grad_norm": 5.7377824783325195,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 4.2849,
      "mean_token_accuracy": 0.7926695197820663,
      "step": 137
    },
    {
      "epoch": 1.104,
      "grad_norm": 5.895854949951172,
      "learning_rate": 3.656565656565657e-05,
      "loss": 3.9685,
      "mean_token_accuracy": 0.8166492581367493,
      "step": 138
    },
    {
      "epoch": 1.112,
      "grad_norm": 6.496169567108154,
      "learning_rate": 3.6464646464646465e-05,
      "loss": 4.2935,
      "mean_token_accuracy": 0.798405259847641,
      "step": 139
    },
    {
      "epoch": 1.12,
      "grad_norm": 5.925403594970703,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 3.5502,
      "mean_token_accuracy": 0.8250329345464706,
      "step": 140
    },
    {
      "epoch": 1.1280000000000001,
      "grad_norm": 6.757926940917969,
      "learning_rate": 3.6262626262626264e-05,
      "loss": 4.0215,
      "mean_token_accuracy": 0.8052553087472916,
      "step": 141
    },
    {
      "epoch": 1.1360000000000001,
      "grad_norm": 5.971244812011719,
      "learning_rate": 3.616161616161616e-05,
      "loss": 4.1611,
      "mean_token_accuracy": 0.7966379076242447,
      "step": 142
    },
    {
      "epoch": 1.144,
      "grad_norm": 6.020002841949463,
      "learning_rate": 3.606060606060606e-05,
      "loss": 3.7227,
      "mean_token_accuracy": 0.8071051388978958,
      "step": 143
    },
    {
      "epoch": 1.152,
      "grad_norm": 5.251856327056885,
      "learning_rate": 3.595959595959596e-05,
      "loss": 3.8428,
      "mean_token_accuracy": 0.8186961710453033,
      "step": 144
    },
    {
      "epoch": 1.16,
      "grad_norm": 5.363802909851074,
      "learning_rate": 3.5858585858585855e-05,
      "loss": 3.1516,
      "mean_token_accuracy": 0.8421822488307953,
      "step": 145
    },
    {
      "epoch": 1.168,
      "grad_norm": 5.687267780303955,
      "learning_rate": 3.575757575757576e-05,
      "loss": 3.8707,
      "mean_token_accuracy": 0.8026652336120605,
      "step": 146
    },
    {
      "epoch": 1.176,
      "grad_norm": 6.096588134765625,
      "learning_rate": 3.565656565656566e-05,
      "loss": 3.6084,
      "mean_token_accuracy": 0.8240565657615662,
      "step": 147
    },
    {
      "epoch": 1.184,
      "grad_norm": 7.285717487335205,
      "learning_rate": 3.555555555555556e-05,
      "loss": 4.0248,
      "mean_token_accuracy": 0.8036085069179535,
      "step": 148
    },
    {
      "epoch": 1.192,
      "grad_norm": 5.360130310058594,
      "learning_rate": 3.545454545454546e-05,
      "loss": 3.7427,
      "mean_token_accuracy": 0.8152587413787842,
      "step": 149
    },
    {
      "epoch": 1.2,
      "grad_norm": 5.485192775726318,
      "learning_rate": 3.535353535353535e-05,
      "loss": 3.4779,
      "mean_token_accuracy": 0.822798877954483,
      "step": 150
    },
    {
      "epoch": 1.208,
      "grad_norm": 6.339715003967285,
      "learning_rate": 3.525252525252525e-05,
      "loss": 4.64,
      "mean_token_accuracy": 0.7758079469203949,
      "step": 151
    },
    {
      "epoch": 1.216,
      "grad_norm": 5.6540422439575195,
      "learning_rate": 3.515151515151515e-05,
      "loss": 3.5393,
      "mean_token_accuracy": 0.8202964663505554,
      "step": 152
    },
    {
      "epoch": 1.224,
      "grad_norm": 6.171947479248047,
      "learning_rate": 3.505050505050505e-05,
      "loss": 3.9077,
      "mean_token_accuracy": 0.8147079646587372,
      "step": 153
    },
    {
      "epoch": 1.232,
      "grad_norm": 5.978511810302734,
      "learning_rate": 3.494949494949495e-05,
      "loss": 3.9126,
      "mean_token_accuracy": 0.8115392625331879,
      "step": 154
    },
    {
      "epoch": 1.24,
      "grad_norm": 6.21269416809082,
      "learning_rate": 3.484848484848485e-05,
      "loss": 3.5828,
      "mean_token_accuracy": 0.8218794912099838,
      "step": 155
    },
    {
      "epoch": 1.248,
      "grad_norm": 5.694769859313965,
      "learning_rate": 3.474747474747475e-05,
      "loss": 3.4313,
      "mean_token_accuracy": 0.8210793286561966,
      "step": 156
    },
    {
      "epoch": 1.256,
      "grad_norm": 5.795802116394043,
      "learning_rate": 3.464646464646465e-05,
      "loss": 3.601,
      "mean_token_accuracy": 0.8306063264608383,
      "step": 157
    },
    {
      "epoch": 1.264,
      "grad_norm": 6.824512004852295,
      "learning_rate": 3.454545454545455e-05,
      "loss": 4.2305,
      "mean_token_accuracy": 0.8084036558866501,
      "step": 158
    },
    {
      "epoch": 1.272,
      "grad_norm": 5.593602180480957,
      "learning_rate": 3.444444444444445e-05,
      "loss": 4.0785,
      "mean_token_accuracy": 0.8063912093639374,
      "step": 159
    },
    {
      "epoch": 1.28,
      "grad_norm": 6.345674991607666,
      "learning_rate": 3.434343434343435e-05,
      "loss": 3.7281,
      "mean_token_accuracy": 0.8293210566043854,
      "step": 160
    },
    {
      "epoch": 1.288,
      "grad_norm": 5.772035598754883,
      "learning_rate": 3.424242424242424e-05,
      "loss": 3.9053,
      "mean_token_accuracy": 0.8084607124328613,
      "step": 161
    },
    {
      "epoch": 1.296,
      "grad_norm": 5.2538042068481445,
      "learning_rate": 3.414141414141414e-05,
      "loss": 3.5784,
      "mean_token_accuracy": 0.8259472250938416,
      "step": 162
    },
    {
      "epoch": 1.304,
      "grad_norm": 6.4647016525268555,
      "learning_rate": 3.4040404040404045e-05,
      "loss": 3.6876,
      "mean_token_accuracy": 0.8131130933761597,
      "step": 163
    },
    {
      "epoch": 1.312,
      "grad_norm": 5.747474670410156,
      "learning_rate": 3.3939393939393945e-05,
      "loss": 3.4403,
      "mean_token_accuracy": 0.8292675763368607,
      "step": 164
    },
    {
      "epoch": 1.32,
      "grad_norm": 5.909630298614502,
      "learning_rate": 3.3838383838383844e-05,
      "loss": 4.1141,
      "mean_token_accuracy": 0.7911320775747299,
      "step": 165
    },
    {
      "epoch": 1.328,
      "grad_norm": 5.7181572914123535,
      "learning_rate": 3.373737373737374e-05,
      "loss": 3.5109,
      "mean_token_accuracy": 0.8262276649475098,
      "step": 166
    },
    {
      "epoch": 1.336,
      "grad_norm": 5.75642728805542,
      "learning_rate": 3.3636363636363636e-05,
      "loss": 3.8356,
      "mean_token_accuracy": 0.8160548806190491,
      "step": 167
    },
    {
      "epoch": 1.3439999999999999,
      "grad_norm": 5.626729965209961,
      "learning_rate": 3.3535353535353536e-05,
      "loss": 3.7526,
      "mean_token_accuracy": 0.8073613941669464,
      "step": 168
    },
    {
      "epoch": 1.3519999999999999,
      "grad_norm": 5.47194766998291,
      "learning_rate": 3.3434343434343435e-05,
      "loss": 3.3927,
      "mean_token_accuracy": 0.8345044106245041,
      "step": 169
    },
    {
      "epoch": 1.3599999999999999,
      "grad_norm": 5.935661792755127,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 3.8877,
      "mean_token_accuracy": 0.8028298169374466,
      "step": 170
    },
    {
      "epoch": 1.3679999999999999,
      "grad_norm": 6.1031107902526855,
      "learning_rate": 3.3232323232323234e-05,
      "loss": 3.159,
      "mean_token_accuracy": 0.8418579548597336,
      "step": 171
    },
    {
      "epoch": 1.376,
      "grad_norm": 5.29210090637207,
      "learning_rate": 3.3131313131313134e-05,
      "loss": 3.4754,
      "mean_token_accuracy": 0.8293817788362503,
      "step": 172
    },
    {
      "epoch": 1.384,
      "grad_norm": 5.36245059967041,
      "learning_rate": 3.303030303030303e-05,
      "loss": 3.5582,
      "mean_token_accuracy": 0.8308709412813187,
      "step": 173
    },
    {
      "epoch": 1.392,
      "grad_norm": 5.807570457458496,
      "learning_rate": 3.292929292929293e-05,
      "loss": 3.6899,
      "mean_token_accuracy": 0.809835895895958,
      "step": 174
    },
    {
      "epoch": 1.4,
      "grad_norm": 6.17417573928833,
      "learning_rate": 3.282828282828283e-05,
      "loss": 4.438,
      "mean_token_accuracy": 0.7785096615552902,
      "step": 175
    },
    {
      "epoch": 1.408,
      "grad_norm": 5.818863868713379,
      "learning_rate": 3.272727272727273e-05,
      "loss": 4.1754,
      "mean_token_accuracy": 0.7955872565507889,
      "step": 176
    },
    {
      "epoch": 1.416,
      "grad_norm": 7.145393371582031,
      "learning_rate": 3.2626262626262624e-05,
      "loss": 3.9973,
      "mean_token_accuracy": 0.7989909946918488,
      "step": 177
    },
    {
      "epoch": 1.424,
      "grad_norm": 5.308239459991455,
      "learning_rate": 3.2525252525252524e-05,
      "loss": 3.8291,
      "mean_token_accuracy": 0.8216778337955475,
      "step": 178
    },
    {
      "epoch": 1.432,
      "grad_norm": 7.332181930541992,
      "learning_rate": 3.2424242424242423e-05,
      "loss": 3.9109,
      "mean_token_accuracy": 0.8071533888578415,
      "step": 179
    },
    {
      "epoch": 1.44,
      "grad_norm": 5.38962984085083,
      "learning_rate": 3.232323232323233e-05,
      "loss": 3.9099,
      "mean_token_accuracy": 0.8167336881160736,
      "step": 180
    },
    {
      "epoch": 1.448,
      "grad_norm": 6.646029949188232,
      "learning_rate": 3.222222222222223e-05,
      "loss": 4.041,
      "mean_token_accuracy": 0.8054783195257187,
      "step": 181
    },
    {
      "epoch": 1.456,
      "grad_norm": 5.449068069458008,
      "learning_rate": 3.212121212121212e-05,
      "loss": 3.8969,
      "mean_token_accuracy": 0.8104897290468216,
      "step": 182
    },
    {
      "epoch": 1.464,
      "grad_norm": 6.279653072357178,
      "learning_rate": 3.202020202020202e-05,
      "loss": 3.8059,
      "mean_token_accuracy": 0.8105663061141968,
      "step": 183
    },
    {
      "epoch": 1.472,
      "grad_norm": 6.5596537590026855,
      "learning_rate": 3.191919191919192e-05,
      "loss": 4.1534,
      "mean_token_accuracy": 0.7909954190254211,
      "step": 184
    },
    {
      "epoch": 1.48,
      "grad_norm": 5.6107354164123535,
      "learning_rate": 3.181818181818182e-05,
      "loss": 3.7084,
      "mean_token_accuracy": 0.8198148310184479,
      "step": 185
    },
    {
      "epoch": 1.488,
      "grad_norm": 5.935013294219971,
      "learning_rate": 3.171717171717172e-05,
      "loss": 3.6008,
      "mean_token_accuracy": 0.8179962188005447,
      "step": 186
    },
    {
      "epoch": 1.496,
      "grad_norm": 6.108509540557861,
      "learning_rate": 3.161616161616161e-05,
      "loss": 4.2591,
      "mean_token_accuracy": 0.802385538816452,
      "step": 187
    },
    {
      "epoch": 1.504,
      "grad_norm": 5.757572650909424,
      "learning_rate": 3.151515151515151e-05,
      "loss": 3.1575,
      "mean_token_accuracy": 0.8446398079395294,
      "step": 188
    },
    {
      "epoch": 1.512,
      "grad_norm": 5.682815074920654,
      "learning_rate": 3.141414141414142e-05,
      "loss": 3.8183,
      "mean_token_accuracy": 0.8149935752153397,
      "step": 189
    },
    {
      "epoch": 1.52,
      "grad_norm": 6.077356338500977,
      "learning_rate": 3.131313131313132e-05,
      "loss": 3.6911,
      "mean_token_accuracy": 0.8199481070041656,
      "step": 190
    },
    {
      "epoch": 1.528,
      "grad_norm": 5.448940277099609,
      "learning_rate": 3.121212121212122e-05,
      "loss": 3.5303,
      "mean_token_accuracy": 0.8298965841531754,
      "step": 191
    },
    {
      "epoch": 1.536,
      "grad_norm": 5.414105415344238,
      "learning_rate": 3.111111111111111e-05,
      "loss": 3.8164,
      "mean_token_accuracy": 0.8245872855186462,
      "step": 192
    },
    {
      "epoch": 1.544,
      "grad_norm": 5.119867324829102,
      "learning_rate": 3.101010101010101e-05,
      "loss": 3.3991,
      "mean_token_accuracy": 0.8331074118614197,
      "step": 193
    },
    {
      "epoch": 1.552,
      "grad_norm": 5.911476135253906,
      "learning_rate": 3.090909090909091e-05,
      "loss": 3.6991,
      "mean_token_accuracy": 0.8288981467485428,
      "step": 194
    },
    {
      "epoch": 1.56,
      "grad_norm": 6.588876247406006,
      "learning_rate": 3.080808080808081e-05,
      "loss": 3.9679,
      "mean_token_accuracy": 0.8007670193910599,
      "step": 195
    },
    {
      "epoch": 1.568,
      "grad_norm": 6.243876934051514,
      "learning_rate": 3.070707070707071e-05,
      "loss": 3.2798,
      "mean_token_accuracy": 0.8313741534948349,
      "step": 196
    },
    {
      "epoch": 1.576,
      "grad_norm": 5.5765180587768555,
      "learning_rate": 3.060606060606061e-05,
      "loss": 3.5246,
      "mean_token_accuracy": 0.8275876641273499,
      "step": 197
    },
    {
      "epoch": 1.584,
      "grad_norm": 5.727989673614502,
      "learning_rate": 3.050505050505051e-05,
      "loss": 3.577,
      "mean_token_accuracy": 0.8309095203876495,
      "step": 198
    },
    {
      "epoch": 1.592,
      "grad_norm": 6.035944938659668,
      "learning_rate": 3.0404040404040406e-05,
      "loss": 3.9191,
      "mean_token_accuracy": 0.8098112493753433,
      "step": 199
    },
    {
      "epoch": 1.6,
      "grad_norm": 6.832411289215088,
      "learning_rate": 3.0303030303030306e-05,
      "loss": 4.0711,
      "mean_token_accuracy": 0.7989842146635056,
      "step": 200
    },
    {
      "epoch": 1.608,
      "grad_norm": 6.298995494842529,
      "learning_rate": 3.0202020202020205e-05,
      "loss": 3.8327,
      "mean_token_accuracy": 0.8034229874610901,
      "step": 201
    },
    {
      "epoch": 1.616,
      "grad_norm": 6.169890880584717,
      "learning_rate": 3.01010101010101e-05,
      "loss": 3.4308,
      "mean_token_accuracy": 0.8316022306680679,
      "step": 202
    },
    {
      "epoch": 1.624,
      "grad_norm": 6.228781700134277,
      "learning_rate": 3e-05,
      "loss": 4.0449,
      "mean_token_accuracy": 0.8119410276412964,
      "step": 203
    },
    {
      "epoch": 1.6320000000000001,
      "grad_norm": 5.267385959625244,
      "learning_rate": 2.98989898989899e-05,
      "loss": 3.3364,
      "mean_token_accuracy": 0.8189917951822281,
      "step": 204
    },
    {
      "epoch": 1.6400000000000001,
      "grad_norm": 6.143648624420166,
      "learning_rate": 2.9797979797979796e-05,
      "loss": 3.6751,
      "mean_token_accuracy": 0.8097221404314041,
      "step": 205
    },
    {
      "epoch": 1.6480000000000001,
      "grad_norm": 5.887429714202881,
      "learning_rate": 2.96969696969697e-05,
      "loss": 3.2366,
      "mean_token_accuracy": 0.8330790549516678,
      "step": 206
    },
    {
      "epoch": 1.6560000000000001,
      "grad_norm": 6.628138065338135,
      "learning_rate": 2.95959595959596e-05,
      "loss": 3.4484,
      "mean_token_accuracy": 0.825881227850914,
      "step": 207
    },
    {
      "epoch": 1.6640000000000001,
      "grad_norm": 6.397181034088135,
      "learning_rate": 2.9494949494949498e-05,
      "loss": 4.1198,
      "mean_token_accuracy": 0.801177367568016,
      "step": 208
    },
    {
      "epoch": 1.6720000000000002,
      "grad_norm": 6.744088649749756,
      "learning_rate": 2.9393939393939394e-05,
      "loss": 3.7321,
      "mean_token_accuracy": 0.8189053982496262,
      "step": 209
    },
    {
      "epoch": 1.6800000000000002,
      "grad_norm": 6.219143390655518,
      "learning_rate": 2.9292929292929294e-05,
      "loss": 3.8731,
      "mean_token_accuracy": 0.807484045624733,
      "step": 210
    },
    {
      "epoch": 1.688,
      "grad_norm": 5.3475751876831055,
      "learning_rate": 2.9191919191919193e-05,
      "loss": 3.3026,
      "mean_token_accuracy": 0.8284660577774048,
      "step": 211
    },
    {
      "epoch": 1.696,
      "grad_norm": 6.080757141113281,
      "learning_rate": 2.909090909090909e-05,
      "loss": 3.9731,
      "mean_token_accuracy": 0.8116171061992645,
      "step": 212
    },
    {
      "epoch": 1.704,
      "grad_norm": 5.819918155670166,
      "learning_rate": 2.898989898989899e-05,
      "loss": 3.146,
      "mean_token_accuracy": 0.83076611161232,
      "step": 213
    },
    {
      "epoch": 1.712,
      "grad_norm": 6.442020416259766,
      "learning_rate": 2.8888888888888888e-05,
      "loss": 3.1441,
      "mean_token_accuracy": 0.834155797958374,
      "step": 214
    },
    {
      "epoch": 1.72,
      "grad_norm": 6.770529270172119,
      "learning_rate": 2.878787878787879e-05,
      "loss": 3.701,
      "mean_token_accuracy": 0.8147666454315186,
      "step": 215
    },
    {
      "epoch": 1.728,
      "grad_norm": 6.345605373382568,
      "learning_rate": 2.868686868686869e-05,
      "loss": 3.4011,
      "mean_token_accuracy": 0.8231780230998993,
      "step": 216
    },
    {
      "epoch": 1.736,
      "grad_norm": 6.269248008728027,
      "learning_rate": 2.8585858585858587e-05,
      "loss": 3.3734,
      "mean_token_accuracy": 0.8338766843080521,
      "step": 217
    },
    {
      "epoch": 1.744,
      "grad_norm": 6.861104965209961,
      "learning_rate": 2.8484848484848486e-05,
      "loss": 3.6933,
      "mean_token_accuracy": 0.8223045021295547,
      "step": 218
    },
    {
      "epoch": 1.752,
      "grad_norm": 5.622055530548096,
      "learning_rate": 2.8383838383838386e-05,
      "loss": 3.6327,
      "mean_token_accuracy": 0.8273986279964447,
      "step": 219
    },
    {
      "epoch": 1.76,
      "grad_norm": 6.05256462097168,
      "learning_rate": 2.8282828282828282e-05,
      "loss": 3.9795,
      "mean_token_accuracy": 0.8149670958518982,
      "step": 220
    },
    {
      "epoch": 1.768,
      "grad_norm": 6.203214645385742,
      "learning_rate": 2.818181818181818e-05,
      "loss": 3.8234,
      "mean_token_accuracy": 0.8010188341140747,
      "step": 221
    },
    {
      "epoch": 1.776,
      "grad_norm": 5.9735798835754395,
      "learning_rate": 2.808080808080808e-05,
      "loss": 3.6441,
      "mean_token_accuracy": 0.8297218382358551,
      "step": 222
    },
    {
      "epoch": 1.784,
      "grad_norm": 5.534307956695557,
      "learning_rate": 2.7979797979797984e-05,
      "loss": 3.9049,
      "mean_token_accuracy": 0.808393806219101,
      "step": 223
    },
    {
      "epoch": 1.792,
      "grad_norm": 5.976230144500732,
      "learning_rate": 2.7878787878787883e-05,
      "loss": 3.1354,
      "mean_token_accuracy": 0.8418057858943939,
      "step": 224
    },
    {
      "epoch": 1.8,
      "grad_norm": 6.141560077667236,
      "learning_rate": 2.777777777777778e-05,
      "loss": 3.9247,
      "mean_token_accuracy": 0.8154790848493576,
      "step": 225
    },
    {
      "epoch": 1.808,
      "grad_norm": 6.0011773109436035,
      "learning_rate": 2.767676767676768e-05,
      "loss": 3.9032,
      "mean_token_accuracy": 0.8127383142709732,
      "step": 226
    },
    {
      "epoch": 1.8159999999999998,
      "grad_norm": 6.12231969833374,
      "learning_rate": 2.7575757575757578e-05,
      "loss": 3.1854,
      "mean_token_accuracy": 0.8354455977678299,
      "step": 227
    },
    {
      "epoch": 1.8239999999999998,
      "grad_norm": 5.492841720581055,
      "learning_rate": 2.7474747474747474e-05,
      "loss": 3.2232,
      "mean_token_accuracy": 0.8330144584178925,
      "step": 228
    },
    {
      "epoch": 1.8319999999999999,
      "grad_norm": 5.566941738128662,
      "learning_rate": 2.7373737373737374e-05,
      "loss": 3.4351,
      "mean_token_accuracy": 0.8365740329027176,
      "step": 229
    },
    {
      "epoch": 1.8399999999999999,
      "grad_norm": 6.055516719818115,
      "learning_rate": 2.7272727272727273e-05,
      "loss": 3.5741,
      "mean_token_accuracy": 0.819349929690361,
      "step": 230
    },
    {
      "epoch": 1.8479999999999999,
      "grad_norm": 5.578907489776611,
      "learning_rate": 2.717171717171717e-05,
      "loss": 3.4948,
      "mean_token_accuracy": 0.8307986855506897,
      "step": 231
    },
    {
      "epoch": 1.8559999999999999,
      "grad_norm": 6.451387882232666,
      "learning_rate": 2.7070707070707075e-05,
      "loss": 3.7951,
      "mean_token_accuracy": 0.8153550177812576,
      "step": 232
    },
    {
      "epoch": 1.8639999999999999,
      "grad_norm": 6.489166736602783,
      "learning_rate": 2.696969696969697e-05,
      "loss": 4.1551,
      "mean_token_accuracy": 0.8079245835542679,
      "step": 233
    },
    {
      "epoch": 1.8719999999999999,
      "grad_norm": 6.350606441497803,
      "learning_rate": 2.686868686868687e-05,
      "loss": 4.2089,
      "mean_token_accuracy": 0.7974926680326462,
      "step": 234
    },
    {
      "epoch": 1.88,
      "grad_norm": 6.2156877517700195,
      "learning_rate": 2.676767676767677e-05,
      "loss": 3.4177,
      "mean_token_accuracy": 0.8269297033548355,
      "step": 235
    },
    {
      "epoch": 1.888,
      "grad_norm": 6.369142532348633,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 3.7993,
      "mean_token_accuracy": 0.8191967755556107,
      "step": 236
    },
    {
      "epoch": 1.896,
      "grad_norm": 5.774569988250732,
      "learning_rate": 2.6565656565656566e-05,
      "loss": 3.7034,
      "mean_token_accuracy": 0.8201487958431244,
      "step": 237
    },
    {
      "epoch": 1.904,
      "grad_norm": 7.003572940826416,
      "learning_rate": 2.6464646464646466e-05,
      "loss": 3.7289,
      "mean_token_accuracy": 0.8157116621732712,
      "step": 238
    },
    {
      "epoch": 1.912,
      "grad_norm": 5.5132036209106445,
      "learning_rate": 2.636363636363636e-05,
      "loss": 3.0822,
      "mean_token_accuracy": 0.8483001440763474,
      "step": 239
    },
    {
      "epoch": 1.92,
      "grad_norm": 5.702081680297852,
      "learning_rate": 2.6262626262626268e-05,
      "loss": 3.3978,
      "mean_token_accuracy": 0.8383439779281616,
      "step": 240
    },
    {
      "epoch": 1.928,
      "grad_norm": 5.952939987182617,
      "learning_rate": 2.6161616161616164e-05,
      "loss": 3.7917,
      "mean_token_accuracy": 0.8186378180980682,
      "step": 241
    },
    {
      "epoch": 1.936,
      "grad_norm": 5.806432247161865,
      "learning_rate": 2.6060606060606063e-05,
      "loss": 3.3405,
      "mean_token_accuracy": 0.8411876261234283,
      "step": 242
    },
    {
      "epoch": 1.944,
      "grad_norm": 5.565011501312256,
      "learning_rate": 2.5959595959595963e-05,
      "loss": 3.7156,
      "mean_token_accuracy": 0.8047986626625061,
      "step": 243
    },
    {
      "epoch": 1.952,
      "grad_norm": 6.6874494552612305,
      "learning_rate": 2.585858585858586e-05,
      "loss": 4.1427,
      "mean_token_accuracy": 0.7969915717840195,
      "step": 244
    },
    {
      "epoch": 1.96,
      "grad_norm": 6.274991989135742,
      "learning_rate": 2.575757575757576e-05,
      "loss": 3.5603,
      "mean_token_accuracy": 0.8269449025392532,
      "step": 245
    },
    {
      "epoch": 1.968,
      "grad_norm": 5.758999824523926,
      "learning_rate": 2.5656565656565658e-05,
      "loss": 2.9123,
      "mean_token_accuracy": 0.8435051888227463,
      "step": 246
    },
    {
      "epoch": 1.976,
      "grad_norm": 6.024221420288086,
      "learning_rate": 2.5555555555555554e-05,
      "loss": 3.9897,
      "mean_token_accuracy": 0.8267959505319595,
      "step": 247
    },
    {
      "epoch": 1.984,
      "grad_norm": 5.789820671081543,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 3.3965,
      "mean_token_accuracy": 0.8389037996530533,
      "step": 248
    },
    {
      "epoch": 1.992,
      "grad_norm": 5.86129903793335,
      "learning_rate": 2.5353535353535356e-05,
      "loss": 3.9343,
      "mean_token_accuracy": 0.8126579076051712,
      "step": 249
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.172614097595215,
      "learning_rate": 2.5252525252525256e-05,
      "loss": 3.2032,
      "mean_token_accuracy": 0.8359029293060303,
      "step": 250
    },
    {
      "epoch": 2.008,
      "grad_norm": 6.1252899169921875,
      "learning_rate": 2.5151515151515155e-05,
      "loss": 3.006,
      "mean_token_accuracy": 0.8488074690103531,
      "step": 251
    },
    {
      "epoch": 2.016,
      "grad_norm": 5.982184410095215,
      "learning_rate": 2.505050505050505e-05,
      "loss": 3.2548,
      "mean_token_accuracy": 0.8324020653963089,
      "step": 252
    },
    {
      "epoch": 2.024,
      "grad_norm": 5.628816604614258,
      "learning_rate": 2.494949494949495e-05,
      "loss": 3.321,
      "mean_token_accuracy": 0.8334167748689651,
      "step": 253
    },
    {
      "epoch": 2.032,
      "grad_norm": 5.734603404998779,
      "learning_rate": 2.4848484848484847e-05,
      "loss": 2.9638,
      "mean_token_accuracy": 0.8483208119869232,
      "step": 254
    },
    {
      "epoch": 2.04,
      "grad_norm": 5.817469120025635,
      "learning_rate": 2.474747474747475e-05,
      "loss": 3.0251,
      "mean_token_accuracy": 0.837944746017456,
      "step": 255
    },
    {
      "epoch": 2.048,
      "grad_norm": 5.197110652923584,
      "learning_rate": 2.464646464646465e-05,
      "loss": 3.1505,
      "mean_token_accuracy": 0.8443950265645981,
      "step": 256
    },
    {
      "epoch": 2.056,
      "grad_norm": 5.90143346786499,
      "learning_rate": 2.4545454545454545e-05,
      "loss": 3.456,
      "mean_token_accuracy": 0.8270305395126343,
      "step": 257
    },
    {
      "epoch": 2.064,
      "grad_norm": 6.32747745513916,
      "learning_rate": 2.4444444444444445e-05,
      "loss": 3.0387,
      "mean_token_accuracy": 0.8455974459648132,
      "step": 258
    },
    {
      "epoch": 2.072,
      "grad_norm": 5.210789203643799,
      "learning_rate": 2.4343434343434344e-05,
      "loss": 3.1112,
      "mean_token_accuracy": 0.847496286034584,
      "step": 259
    },
    {
      "epoch": 2.08,
      "grad_norm": 5.2880167961120605,
      "learning_rate": 2.4242424242424244e-05,
      "loss": 2.7432,
      "mean_token_accuracy": 0.8593413680791855,
      "step": 260
    },
    {
      "epoch": 2.088,
      "grad_norm": 5.868934154510498,
      "learning_rate": 2.4141414141414143e-05,
      "loss": 3.1459,
      "mean_token_accuracy": 0.8433663100004196,
      "step": 261
    },
    {
      "epoch": 2.096,
      "grad_norm": 5.824329376220703,
      "learning_rate": 2.404040404040404e-05,
      "loss": 3.5007,
      "mean_token_accuracy": 0.8235083371400833,
      "step": 262
    },
    {
      "epoch": 2.104,
      "grad_norm": 5.528572082519531,
      "learning_rate": 2.393939393939394e-05,
      "loss": 2.854,
      "mean_token_accuracy": 0.8565276563167572,
      "step": 263
    },
    {
      "epoch": 2.112,
      "grad_norm": 6.147289752960205,
      "learning_rate": 2.3838383838383842e-05,
      "loss": 2.9618,
      "mean_token_accuracy": 0.8526737242937088,
      "step": 264
    },
    {
      "epoch": 2.12,
      "grad_norm": 5.65183687210083,
      "learning_rate": 2.3737373737373738e-05,
      "loss": 3.0466,
      "mean_token_accuracy": 0.8418086171150208,
      "step": 265
    },
    {
      "epoch": 2.128,
      "grad_norm": 6.315326690673828,
      "learning_rate": 2.3636363636363637e-05,
      "loss": 3.4835,
      "mean_token_accuracy": 0.8310810178518295,
      "step": 266
    },
    {
      "epoch": 2.136,
      "grad_norm": 5.648726463317871,
      "learning_rate": 2.3535353535353537e-05,
      "loss": 3.6856,
      "mean_token_accuracy": 0.8161235004663467,
      "step": 267
    },
    {
      "epoch": 2.144,
      "grad_norm": 5.972370624542236,
      "learning_rate": 2.3434343434343436e-05,
      "loss": 3.0239,
      "mean_token_accuracy": 0.8378438502550125,
      "step": 268
    },
    {
      "epoch": 2.152,
      "grad_norm": 5.596478462219238,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 3.0652,
      "mean_token_accuracy": 0.8487441837787628,
      "step": 269
    },
    {
      "epoch": 2.16,
      "grad_norm": 6.5623908042907715,
      "learning_rate": 2.3232323232323232e-05,
      "loss": 3.1893,
      "mean_token_accuracy": 0.834345743060112,
      "step": 270
    },
    {
      "epoch": 2.168,
      "grad_norm": 6.067134380340576,
      "learning_rate": 2.313131313131313e-05,
      "loss": 3.0515,
      "mean_token_accuracy": 0.8387996703386307,
      "step": 271
    },
    {
      "epoch": 2.176,
      "grad_norm": 6.862771987915039,
      "learning_rate": 2.3030303030303034e-05,
      "loss": 3.2861,
      "mean_token_accuracy": 0.8361453711986542,
      "step": 272
    },
    {
      "epoch": 2.184,
      "grad_norm": 6.166702747344971,
      "learning_rate": 2.292929292929293e-05,
      "loss": 3.6596,
      "mean_token_accuracy": 0.813729852437973,
      "step": 273
    },
    {
      "epoch": 2.192,
      "grad_norm": 5.386834621429443,
      "learning_rate": 2.282828282828283e-05,
      "loss": 3.4533,
      "mean_token_accuracy": 0.8274287581443787,
      "step": 274
    },
    {
      "epoch": 2.2,
      "grad_norm": 5.261857986450195,
      "learning_rate": 2.272727272727273e-05,
      "loss": 2.7643,
      "mean_token_accuracy": 0.8519167453050613,
      "step": 275
    },
    {
      "epoch": 2.208,
      "grad_norm": 5.241461753845215,
      "learning_rate": 2.262626262626263e-05,
      "loss": 3.051,
      "mean_token_accuracy": 0.8460838198661804,
      "step": 276
    },
    {
      "epoch": 2.216,
      "grad_norm": 5.825733184814453,
      "learning_rate": 2.2525252525252528e-05,
      "loss": 3.2424,
      "mean_token_accuracy": 0.8403096050024033,
      "step": 277
    },
    {
      "epoch": 2.224,
      "grad_norm": 5.554264545440674,
      "learning_rate": 2.2424242424242424e-05,
      "loss": 3.1676,
      "mean_token_accuracy": 0.8293324261903763,
      "step": 278
    },
    {
      "epoch": 2.232,
      "grad_norm": 5.81028413772583,
      "learning_rate": 2.2323232323232324e-05,
      "loss": 3.184,
      "mean_token_accuracy": 0.8394870609045029,
      "step": 279
    },
    {
      "epoch": 2.24,
      "grad_norm": 6.055649280548096,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 3.0117,
      "mean_token_accuracy": 0.850115180015564,
      "step": 280
    },
    {
      "epoch": 2.248,
      "grad_norm": 5.307283401489258,
      "learning_rate": 2.2121212121212123e-05,
      "loss": 2.9076,
      "mean_token_accuracy": 0.8414607346057892,
      "step": 281
    },
    {
      "epoch": 2.2560000000000002,
      "grad_norm": 5.909352779388428,
      "learning_rate": 2.2020202020202022e-05,
      "loss": 3.4122,
      "mean_token_accuracy": 0.8330782055854797,
      "step": 282
    },
    {
      "epoch": 2.2640000000000002,
      "grad_norm": 5.730319499969482,
      "learning_rate": 2.191919191919192e-05,
      "loss": 3.2416,
      "mean_token_accuracy": 0.83427894115448,
      "step": 283
    },
    {
      "epoch": 2.2720000000000002,
      "grad_norm": 6.638869762420654,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 2.8864,
      "mean_token_accuracy": 0.8467890173196793,
      "step": 284
    },
    {
      "epoch": 2.2800000000000002,
      "grad_norm": 6.180927753448486,
      "learning_rate": 2.171717171717172e-05,
      "loss": 3.3418,
      "mean_token_accuracy": 0.8295275717973709,
      "step": 285
    },
    {
      "epoch": 2.288,
      "grad_norm": 5.50770902633667,
      "learning_rate": 2.1616161616161617e-05,
      "loss": 3.3557,
      "mean_token_accuracy": 0.8279571235179901,
      "step": 286
    },
    {
      "epoch": 2.296,
      "grad_norm": 6.085552215576172,
      "learning_rate": 2.1515151515151516e-05,
      "loss": 2.7873,
      "mean_token_accuracy": 0.8584895879030228,
      "step": 287
    },
    {
      "epoch": 2.304,
      "grad_norm": 5.311261177062988,
      "learning_rate": 2.1414141414141416e-05,
      "loss": 2.786,
      "mean_token_accuracy": 0.8529232293367386,
      "step": 288
    },
    {
      "epoch": 2.312,
      "grad_norm": 6.482555389404297,
      "learning_rate": 2.1313131313131315e-05,
      "loss": 3.3625,
      "mean_token_accuracy": 0.8343013226985931,
      "step": 289
    },
    {
      "epoch": 2.32,
      "grad_norm": 5.101843357086182,
      "learning_rate": 2.1212121212121215e-05,
      "loss": 3.0018,
      "mean_token_accuracy": 0.8565692156553268,
      "step": 290
    },
    {
      "epoch": 2.328,
      "grad_norm": 6.496617317199707,
      "learning_rate": 2.111111111111111e-05,
      "loss": 3.5081,
      "mean_token_accuracy": 0.8281321227550507,
      "step": 291
    },
    {
      "epoch": 2.336,
      "grad_norm": 6.459492206573486,
      "learning_rate": 2.101010101010101e-05,
      "loss": 3.2684,
      "mean_token_accuracy": 0.8464506566524506,
      "step": 292
    },
    {
      "epoch": 2.344,
      "grad_norm": 6.296543121337891,
      "learning_rate": 2.090909090909091e-05,
      "loss": 2.9787,
      "mean_token_accuracy": 0.8391922265291214,
      "step": 293
    },
    {
      "epoch": 2.352,
      "grad_norm": 5.939156532287598,
      "learning_rate": 2.080808080808081e-05,
      "loss": 3.2943,
      "mean_token_accuracy": 0.8347084373235703,
      "step": 294
    },
    {
      "epoch": 2.36,
      "grad_norm": 5.988732814788818,
      "learning_rate": 2.070707070707071e-05,
      "loss": 3.1281,
      "mean_token_accuracy": 0.8483212292194366,
      "step": 295
    },
    {
      "epoch": 2.368,
      "grad_norm": 7.110536098480225,
      "learning_rate": 2.0606060606060608e-05,
      "loss": 3.1449,
      "mean_token_accuracy": 0.8350488543510437,
      "step": 296
    },
    {
      "epoch": 2.376,
      "grad_norm": 6.519949436187744,
      "learning_rate": 2.0505050505050504e-05,
      "loss": 2.6449,
      "mean_token_accuracy": 0.8593619167804718,
      "step": 297
    },
    {
      "epoch": 2.384,
      "grad_norm": 5.815298557281494,
      "learning_rate": 2.0404040404040407e-05,
      "loss": 3.0911,
      "mean_token_accuracy": 0.837956115603447,
      "step": 298
    },
    {
      "epoch": 2.392,
      "grad_norm": 5.741540908813477,
      "learning_rate": 2.0303030303030303e-05,
      "loss": 3.1501,
      "mean_token_accuracy": 0.8409707248210907,
      "step": 299
    },
    {
      "epoch": 2.4,
      "grad_norm": 5.991777420043945,
      "learning_rate": 2.0202020202020203e-05,
      "loss": 2.9707,
      "mean_token_accuracy": 0.858299732208252,
      "step": 300
    },
    {
      "epoch": 2.408,
      "grad_norm": 6.3398847579956055,
      "learning_rate": 2.0101010101010102e-05,
      "loss": 3.4141,
      "mean_token_accuracy": 0.830435261130333,
      "step": 301
    },
    {
      "epoch": 2.416,
      "grad_norm": 6.837295055389404,
      "learning_rate": 2e-05,
      "loss": 3.3923,
      "mean_token_accuracy": 0.8256559520959854,
      "step": 302
    },
    {
      "epoch": 2.424,
      "grad_norm": 5.505060195922852,
      "learning_rate": 1.98989898989899e-05,
      "loss": 3.0673,
      "mean_token_accuracy": 0.8478615581989288,
      "step": 303
    },
    {
      "epoch": 2.432,
      "grad_norm": 6.913103103637695,
      "learning_rate": 1.9797979797979797e-05,
      "loss": 3.5381,
      "mean_token_accuracy": 0.8223972916603088,
      "step": 304
    },
    {
      "epoch": 2.44,
      "grad_norm": 6.902682304382324,
      "learning_rate": 1.9696969696969697e-05,
      "loss": 3.598,
      "mean_token_accuracy": 0.8203122019767761,
      "step": 305
    },
    {
      "epoch": 2.448,
      "grad_norm": 5.36390495300293,
      "learning_rate": 1.95959595959596e-05,
      "loss": 3.1101,
      "mean_token_accuracy": 0.8461343050003052,
      "step": 306
    },
    {
      "epoch": 2.456,
      "grad_norm": 6.57292366027832,
      "learning_rate": 1.9494949494949496e-05,
      "loss": 3.6492,
      "mean_token_accuracy": 0.8245173096656799,
      "step": 307
    },
    {
      "epoch": 2.464,
      "grad_norm": 5.893022537231445,
      "learning_rate": 1.9393939393939395e-05,
      "loss": 2.7866,
      "mean_token_accuracy": 0.8550339192152023,
      "step": 308
    },
    {
      "epoch": 2.472,
      "grad_norm": 5.711099624633789,
      "learning_rate": 1.9292929292929295e-05,
      "loss": 3.0009,
      "mean_token_accuracy": 0.8481545150279999,
      "step": 309
    },
    {
      "epoch": 2.48,
      "grad_norm": 6.395712375640869,
      "learning_rate": 1.919191919191919e-05,
      "loss": 2.8936,
      "mean_token_accuracy": 0.8416745364665985,
      "step": 310
    },
    {
      "epoch": 2.488,
      "grad_norm": 5.5052289962768555,
      "learning_rate": 1.9090909090909094e-05,
      "loss": 2.9872,
      "mean_token_accuracy": 0.8463147729635239,
      "step": 311
    },
    {
      "epoch": 2.496,
      "grad_norm": 6.273165702819824,
      "learning_rate": 1.898989898989899e-05,
      "loss": 2.7584,
      "mean_token_accuracy": 0.8467618376016617,
      "step": 312
    },
    {
      "epoch": 2.504,
      "grad_norm": 5.292394161224365,
      "learning_rate": 1.888888888888889e-05,
      "loss": 2.9572,
      "mean_token_accuracy": 0.8431327790021896,
      "step": 313
    },
    {
      "epoch": 2.512,
      "grad_norm": 5.991566181182861,
      "learning_rate": 1.878787878787879e-05,
      "loss": 2.9398,
      "mean_token_accuracy": 0.8492739796638489,
      "step": 314
    },
    {
      "epoch": 2.52,
      "grad_norm": 5.635786533355713,
      "learning_rate": 1.8686868686868688e-05,
      "loss": 3.1451,
      "mean_token_accuracy": 0.854804664850235,
      "step": 315
    },
    {
      "epoch": 2.528,
      "grad_norm": 6.714171886444092,
      "learning_rate": 1.8585858585858588e-05,
      "loss": 3.0965,
      "mean_token_accuracy": 0.8423555940389633,
      "step": 316
    },
    {
      "epoch": 2.536,
      "grad_norm": 5.751944065093994,
      "learning_rate": 1.8484848484848487e-05,
      "loss": 3.0769,
      "mean_token_accuracy": 0.8551096171140671,
      "step": 317
    },
    {
      "epoch": 2.544,
      "grad_norm": 6.406241416931152,
      "learning_rate": 1.8383838383838383e-05,
      "loss": 2.8228,
      "mean_token_accuracy": 0.861114576458931,
      "step": 318
    },
    {
      "epoch": 2.552,
      "grad_norm": 6.585404396057129,
      "learning_rate": 1.8282828282828286e-05,
      "loss": 3.2246,
      "mean_token_accuracy": 0.8409701138734818,
      "step": 319
    },
    {
      "epoch": 2.56,
      "grad_norm": 6.6673264503479,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 3.0054,
      "mean_token_accuracy": 0.8525452762842178,
      "step": 320
    },
    {
      "epoch": 2.568,
      "grad_norm": 6.316901206970215,
      "learning_rate": 1.808080808080808e-05,
      "loss": 3.4642,
      "mean_token_accuracy": 0.8289849609136581,
      "step": 321
    },
    {
      "epoch": 2.576,
      "grad_norm": 7.191961288452148,
      "learning_rate": 1.797979797979798e-05,
      "loss": 3.276,
      "mean_token_accuracy": 0.8262393325567245,
      "step": 322
    },
    {
      "epoch": 2.584,
      "grad_norm": 5.743154525756836,
      "learning_rate": 1.787878787878788e-05,
      "loss": 2.6852,
      "mean_token_accuracy": 0.8636340796947479,
      "step": 323
    },
    {
      "epoch": 2.592,
      "grad_norm": 5.567448616027832,
      "learning_rate": 1.777777777777778e-05,
      "loss": 3.0029,
      "mean_token_accuracy": 0.8480563163757324,
      "step": 324
    },
    {
      "epoch": 2.6,
      "grad_norm": 6.558039665222168,
      "learning_rate": 1.7676767676767676e-05,
      "loss": 2.8954,
      "mean_token_accuracy": 0.8490820229053497,
      "step": 325
    },
    {
      "epoch": 2.608,
      "grad_norm": 6.237626075744629,
      "learning_rate": 1.7575757575757576e-05,
      "loss": 3.2673,
      "mean_token_accuracy": 0.8418715596199036,
      "step": 326
    },
    {
      "epoch": 2.616,
      "grad_norm": 6.388242721557617,
      "learning_rate": 1.7474747474747475e-05,
      "loss": 3.5492,
      "mean_token_accuracy": 0.8205874115228653,
      "step": 327
    },
    {
      "epoch": 2.624,
      "grad_norm": 6.261191368103027,
      "learning_rate": 1.7373737373737375e-05,
      "loss": 3.784,
      "mean_token_accuracy": 0.8178833723068237,
      "step": 328
    },
    {
      "epoch": 2.632,
      "grad_norm": 5.890114784240723,
      "learning_rate": 1.7272727272727274e-05,
      "loss": 2.7545,
      "mean_token_accuracy": 0.8488318920135498,
      "step": 329
    },
    {
      "epoch": 2.64,
      "grad_norm": 6.187424182891846,
      "learning_rate": 1.7171717171717173e-05,
      "loss": 3.5939,
      "mean_token_accuracy": 0.827822282910347,
      "step": 330
    },
    {
      "epoch": 2.648,
      "grad_norm": 6.662416458129883,
      "learning_rate": 1.707070707070707e-05,
      "loss": 3.2883,
      "mean_token_accuracy": 0.8301158398389816,
      "step": 331
    },
    {
      "epoch": 2.656,
      "grad_norm": 6.358303546905518,
      "learning_rate": 1.6969696969696972e-05,
      "loss": 3.1986,
      "mean_token_accuracy": 0.8457056134939194,
      "step": 332
    },
    {
      "epoch": 2.664,
      "grad_norm": 5.967710971832275,
      "learning_rate": 1.686868686868687e-05,
      "loss": 2.9175,
      "mean_token_accuracy": 0.8557797521352768,
      "step": 333
    },
    {
      "epoch": 2.672,
      "grad_norm": 5.508559703826904,
      "learning_rate": 1.6767676767676768e-05,
      "loss": 3.1204,
      "mean_token_accuracy": 0.8435854762792587,
      "step": 334
    },
    {
      "epoch": 2.68,
      "grad_norm": 6.057641506195068,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 3.5826,
      "mean_token_accuracy": 0.8250070810317993,
      "step": 335
    },
    {
      "epoch": 2.6879999999999997,
      "grad_norm": 5.788311004638672,
      "learning_rate": 1.6565656565656567e-05,
      "loss": 2.9172,
      "mean_token_accuracy": 0.8496552407741547,
      "step": 336
    },
    {
      "epoch": 2.6959999999999997,
      "grad_norm": 7.819416522979736,
      "learning_rate": 1.6464646464646466e-05,
      "loss": 3.1982,
      "mean_token_accuracy": 0.8477791100740433,
      "step": 337
    },
    {
      "epoch": 2.7039999999999997,
      "grad_norm": 6.290533065795898,
      "learning_rate": 1.6363636363636366e-05,
      "loss": 3.289,
      "mean_token_accuracy": 0.8374428898096085,
      "step": 338
    },
    {
      "epoch": 2.7119999999999997,
      "grad_norm": 6.164359092712402,
      "learning_rate": 1.6262626262626262e-05,
      "loss": 3.0955,
      "mean_token_accuracy": 0.8332884609699249,
      "step": 339
    },
    {
      "epoch": 2.7199999999999998,
      "grad_norm": 6.6753339767456055,
      "learning_rate": 1.6161616161616165e-05,
      "loss": 2.9626,
      "mean_token_accuracy": 0.8437196165323257,
      "step": 340
    },
    {
      "epoch": 2.7279999999999998,
      "grad_norm": 5.542717933654785,
      "learning_rate": 1.606060606060606e-05,
      "loss": 3.2322,
      "mean_token_accuracy": 0.8421852588653564,
      "step": 341
    },
    {
      "epoch": 2.7359999999999998,
      "grad_norm": 5.837277889251709,
      "learning_rate": 1.595959595959596e-05,
      "loss": 3.1454,
      "mean_token_accuracy": 0.8469538539648056,
      "step": 342
    },
    {
      "epoch": 2.7439999999999998,
      "grad_norm": 5.96569299697876,
      "learning_rate": 1.585858585858586e-05,
      "loss": 2.8061,
      "mean_token_accuracy": 0.8568727970123291,
      "step": 343
    },
    {
      "epoch": 2.752,
      "grad_norm": 6.756287574768066,
      "learning_rate": 1.5757575757575756e-05,
      "loss": 2.792,
      "mean_token_accuracy": 0.8376488238573074,
      "step": 344
    },
    {
      "epoch": 2.76,
      "grad_norm": 6.800328254699707,
      "learning_rate": 1.565656565656566e-05,
      "loss": 3.3131,
      "mean_token_accuracy": 0.8389277309179306,
      "step": 345
    },
    {
      "epoch": 2.768,
      "grad_norm": 5.755626201629639,
      "learning_rate": 1.5555555555555555e-05,
      "loss": 2.9014,
      "mean_token_accuracy": 0.8483488708734512,
      "step": 346
    },
    {
      "epoch": 2.776,
      "grad_norm": 5.480167865753174,
      "learning_rate": 1.5454545454545454e-05,
      "loss": 2.8085,
      "mean_token_accuracy": 0.8566093593835831,
      "step": 347
    },
    {
      "epoch": 2.784,
      "grad_norm": 6.205705642700195,
      "learning_rate": 1.5353535353535354e-05,
      "loss": 3.0061,
      "mean_token_accuracy": 0.8438935428857803,
      "step": 348
    },
    {
      "epoch": 2.792,
      "grad_norm": 6.319244384765625,
      "learning_rate": 1.5252525252525255e-05,
      "loss": 2.9846,
      "mean_token_accuracy": 0.8455973863601685,
      "step": 349
    },
    {
      "epoch": 2.8,
      "grad_norm": 5.499710559844971,
      "learning_rate": 1.5151515151515153e-05,
      "loss": 3.0732,
      "mean_token_accuracy": 0.8468042612075806,
      "step": 350
    },
    {
      "epoch": 2.808,
      "grad_norm": 5.381649494171143,
      "learning_rate": 1.505050505050505e-05,
      "loss": 3.4026,
      "mean_token_accuracy": 0.840883806347847,
      "step": 351
    },
    {
      "epoch": 2.816,
      "grad_norm": 5.975292205810547,
      "learning_rate": 1.494949494949495e-05,
      "loss": 3.3919,
      "mean_token_accuracy": 0.8322191834449768,
      "step": 352
    },
    {
      "epoch": 2.824,
      "grad_norm": 6.1554646492004395,
      "learning_rate": 1.484848484848485e-05,
      "loss": 2.9314,
      "mean_token_accuracy": 0.8548919409513474,
      "step": 353
    },
    {
      "epoch": 2.832,
      "grad_norm": 5.832248210906982,
      "learning_rate": 1.4747474747474749e-05,
      "loss": 2.3767,
      "mean_token_accuracy": 0.8693199008703232,
      "step": 354
    },
    {
      "epoch": 2.84,
      "grad_norm": 5.911506175994873,
      "learning_rate": 1.4646464646464647e-05,
      "loss": 2.7502,
      "mean_token_accuracy": 0.8517532050609589,
      "step": 355
    },
    {
      "epoch": 2.848,
      "grad_norm": 6.765252113342285,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 3.6221,
      "mean_token_accuracy": 0.8170515298843384,
      "step": 356
    },
    {
      "epoch": 2.856,
      "grad_norm": 6.607264518737793,
      "learning_rate": 1.4444444444444444e-05,
      "loss": 3.1797,
      "mean_token_accuracy": 0.8341539800167084,
      "step": 357
    },
    {
      "epoch": 2.864,
      "grad_norm": 6.446462154388428,
      "learning_rate": 1.4343434343434345e-05,
      "loss": 2.9366,
      "mean_token_accuracy": 0.8581055998802185,
      "step": 358
    },
    {
      "epoch": 2.872,
      "grad_norm": 6.300421714782715,
      "learning_rate": 1.4242424242424243e-05,
      "loss": 3.3633,
      "mean_token_accuracy": 0.8499312251806259,
      "step": 359
    },
    {
      "epoch": 2.88,
      "grad_norm": 5.600719451904297,
      "learning_rate": 1.4141414141414141e-05,
      "loss": 3.088,
      "mean_token_accuracy": 0.8488198220729828,
      "step": 360
    },
    {
      "epoch": 2.888,
      "grad_norm": 5.980778217315674,
      "learning_rate": 1.404040404040404e-05,
      "loss": 2.9933,
      "mean_token_accuracy": 0.8443039804697037,
      "step": 361
    },
    {
      "epoch": 2.896,
      "grad_norm": 6.331740856170654,
      "learning_rate": 1.3939393939393942e-05,
      "loss": 2.9548,
      "mean_token_accuracy": 0.8464246839284897,
      "step": 362
    },
    {
      "epoch": 2.904,
      "grad_norm": 6.095213890075684,
      "learning_rate": 1.383838383838384e-05,
      "loss": 3.4251,
      "mean_token_accuracy": 0.8375514298677444,
      "step": 363
    },
    {
      "epoch": 2.912,
      "grad_norm": 6.7452239990234375,
      "learning_rate": 1.3737373737373737e-05,
      "loss": 3.3106,
      "mean_token_accuracy": 0.8348345905542374,
      "step": 364
    },
    {
      "epoch": 2.92,
      "grad_norm": 5.934120178222656,
      "learning_rate": 1.3636363636363637e-05,
      "loss": 3.1045,
      "mean_token_accuracy": 0.8491770774126053,
      "step": 365
    },
    {
      "epoch": 2.928,
      "grad_norm": 5.6627912521362305,
      "learning_rate": 1.3535353535353538e-05,
      "loss": 2.9255,
      "mean_token_accuracy": 0.8593737334012985,
      "step": 366
    },
    {
      "epoch": 2.936,
      "grad_norm": 6.51497745513916,
      "learning_rate": 1.3434343434343436e-05,
      "loss": 3.2813,
      "mean_token_accuracy": 0.8357561677694321,
      "step": 367
    },
    {
      "epoch": 2.944,
      "grad_norm": 6.286316394805908,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 2.8042,
      "mean_token_accuracy": 0.8635920882225037,
      "step": 368
    },
    {
      "epoch": 2.952,
      "grad_norm": 6.30144739151001,
      "learning_rate": 1.3232323232323233e-05,
      "loss": 2.7913,
      "mean_token_accuracy": 0.8532185405492783,
      "step": 369
    },
    {
      "epoch": 2.96,
      "grad_norm": 5.6404314041137695,
      "learning_rate": 1.3131313131313134e-05,
      "loss": 2.8048,
      "mean_token_accuracy": 0.8557418137788773,
      "step": 370
    },
    {
      "epoch": 2.968,
      "grad_norm": 6.2682342529296875,
      "learning_rate": 1.3030303030303032e-05,
      "loss": 3.3507,
      "mean_token_accuracy": 0.8299184590578079,
      "step": 371
    },
    {
      "epoch": 2.976,
      "grad_norm": 5.7995219230651855,
      "learning_rate": 1.292929292929293e-05,
      "loss": 2.9806,
      "mean_token_accuracy": 0.8536931127309799,
      "step": 372
    },
    {
      "epoch": 2.984,
      "grad_norm": 6.288918972015381,
      "learning_rate": 1.2828282828282829e-05,
      "loss": 3.1404,
      "mean_token_accuracy": 0.8438131213188171,
      "step": 373
    },
    {
      "epoch": 2.992,
      "grad_norm": 6.12652587890625,
      "learning_rate": 1.2727272727272727e-05,
      "loss": 3.1072,
      "mean_token_accuracy": 0.8395776003599167,
      "step": 374
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.006952285766602,
      "learning_rate": 1.2626262626262628e-05,
      "loss": 3.0308,
      "mean_token_accuracy": 0.8461359292268753,
      "step": 375
    },
    {
      "epoch": 3.008,
      "grad_norm": 5.877007961273193,
      "learning_rate": 1.2525252525252526e-05,
      "loss": 2.7955,
      "mean_token_accuracy": 0.8688687086105347,
      "step": 376
    },
    {
      "epoch": 3.016,
      "grad_norm": 5.722151756286621,
      "learning_rate": 1.2424242424242424e-05,
      "loss": 2.4909,
      "mean_token_accuracy": 0.8692611008882523,
      "step": 377
    },
    {
      "epoch": 3.024,
      "grad_norm": 5.512819290161133,
      "learning_rate": 1.2323232323232325e-05,
      "loss": 2.5139,
      "mean_token_accuracy": 0.8634871691465378,
      "step": 378
    },
    {
      "epoch": 3.032,
      "grad_norm": 5.516348838806152,
      "learning_rate": 1.2222222222222222e-05,
      "loss": 2.8559,
      "mean_token_accuracy": 0.8596174418926239,
      "step": 379
    },
    {
      "epoch": 3.04,
      "grad_norm": 5.508405685424805,
      "learning_rate": 1.2121212121212122e-05,
      "loss": 2.5545,
      "mean_token_accuracy": 0.8677777200937271,
      "step": 380
    },
    {
      "epoch": 3.048,
      "grad_norm": 6.354642868041992,
      "learning_rate": 1.202020202020202e-05,
      "loss": 3.0228,
      "mean_token_accuracy": 0.8556893318891525,
      "step": 381
    },
    {
      "epoch": 3.056,
      "grad_norm": 6.423473834991455,
      "learning_rate": 1.1919191919191921e-05,
      "loss": 3.072,
      "mean_token_accuracy": 0.8429125100374222,
      "step": 382
    },
    {
      "epoch": 3.064,
      "grad_norm": 5.950673580169678,
      "learning_rate": 1.1818181818181819e-05,
      "loss": 3.4449,
      "mean_token_accuracy": 0.8237967044115067,
      "step": 383
    },
    {
      "epoch": 3.072,
      "grad_norm": 6.789544582366943,
      "learning_rate": 1.1717171717171718e-05,
      "loss": 2.8032,
      "mean_token_accuracy": 0.8490213006734848,
      "step": 384
    },
    {
      "epoch": 3.08,
      "grad_norm": 5.8454413414001465,
      "learning_rate": 1.1616161616161616e-05,
      "loss": 2.7207,
      "mean_token_accuracy": 0.8605297058820724,
      "step": 385
    },
    {
      "epoch": 3.088,
      "grad_norm": 6.446214199066162,
      "learning_rate": 1.1515151515151517e-05,
      "loss": 3.2414,
      "mean_token_accuracy": 0.8328104317188263,
      "step": 386
    },
    {
      "epoch": 3.096,
      "grad_norm": 5.578056335449219,
      "learning_rate": 1.1414141414141415e-05,
      "loss": 2.8079,
      "mean_token_accuracy": 0.8580958545207977,
      "step": 387
    },
    {
      "epoch": 3.104,
      "grad_norm": 5.806094169616699,
      "learning_rate": 1.1313131313131314e-05,
      "loss": 2.6591,
      "mean_token_accuracy": 0.8651483207941055,
      "step": 388
    },
    {
      "epoch": 3.112,
      "grad_norm": 4.987120151519775,
      "learning_rate": 1.1212121212121212e-05,
      "loss": 2.2249,
      "mean_token_accuracy": 0.8830093741416931,
      "step": 389
    },
    {
      "epoch": 3.12,
      "grad_norm": 6.069543361663818,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 2.8345,
      "mean_token_accuracy": 0.8573679178953171,
      "step": 390
    },
    {
      "epoch": 3.128,
      "grad_norm": 6.578948974609375,
      "learning_rate": 1.1010101010101011e-05,
      "loss": 3.4663,
      "mean_token_accuracy": 0.8265634626150131,
      "step": 391
    },
    {
      "epoch": 3.136,
      "grad_norm": 6.3034772872924805,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 2.5849,
      "mean_token_accuracy": 0.8636100441217422,
      "step": 392
    },
    {
      "epoch": 3.144,
      "grad_norm": 5.5531206130981445,
      "learning_rate": 1.0808080808080808e-05,
      "loss": 2.8643,
      "mean_token_accuracy": 0.8591476529836655,
      "step": 393
    },
    {
      "epoch": 3.152,
      "grad_norm": 5.342994689941406,
      "learning_rate": 1.0707070707070708e-05,
      "loss": 2.4832,
      "mean_token_accuracy": 0.868215799331665,
      "step": 394
    },
    {
      "epoch": 3.16,
      "grad_norm": 6.020049095153809,
      "learning_rate": 1.0606060606060607e-05,
      "loss": 2.7882,
      "mean_token_accuracy": 0.860662654042244,
      "step": 395
    },
    {
      "epoch": 3.168,
      "grad_norm": 6.296903610229492,
      "learning_rate": 1.0505050505050505e-05,
      "loss": 2.6621,
      "mean_token_accuracy": 0.8558385968208313,
      "step": 396
    },
    {
      "epoch": 3.176,
      "grad_norm": 5.6435980796813965,
      "learning_rate": 1.0404040404040405e-05,
      "loss": 2.5996,
      "mean_token_accuracy": 0.8638162761926651,
      "step": 397
    },
    {
      "epoch": 3.184,
      "grad_norm": 5.6095123291015625,
      "learning_rate": 1.0303030303030304e-05,
      "loss": 2.7824,
      "mean_token_accuracy": 0.8573920726776123,
      "step": 398
    },
    {
      "epoch": 3.192,
      "grad_norm": 5.77184534072876,
      "learning_rate": 1.0202020202020204e-05,
      "loss": 3.2982,
      "mean_token_accuracy": 0.8455280959606171,
      "step": 399
    },
    {
      "epoch": 3.2,
      "grad_norm": 6.490416049957275,
      "learning_rate": 1.0101010101010101e-05,
      "loss": 2.8005,
      "mean_token_accuracy": 0.8535754829645157,
      "step": 400
    },
    {
      "epoch": 3.208,
      "grad_norm": 5.768113613128662,
      "learning_rate": 1e-05,
      "loss": 2.5534,
      "mean_token_accuracy": 0.8649332523345947,
      "step": 401
    },
    {
      "epoch": 3.216,
      "grad_norm": 6.118169784545898,
      "learning_rate": 9.898989898989899e-06,
      "loss": 2.8782,
      "mean_token_accuracy": 0.8546314835548401,
      "step": 402
    },
    {
      "epoch": 3.224,
      "grad_norm": 5.9188456535339355,
      "learning_rate": 9.7979797979798e-06,
      "loss": 2.8495,
      "mean_token_accuracy": 0.8487619459629059,
      "step": 403
    },
    {
      "epoch": 3.232,
      "grad_norm": 6.8953986167907715,
      "learning_rate": 9.696969696969698e-06,
      "loss": 3.2653,
      "mean_token_accuracy": 0.842998206615448,
      "step": 404
    },
    {
      "epoch": 3.24,
      "grad_norm": 6.311713218688965,
      "learning_rate": 9.595959595959595e-06,
      "loss": 3.0002,
      "mean_token_accuracy": 0.8414562940597534,
      "step": 405
    },
    {
      "epoch": 3.248,
      "grad_norm": 6.360631465911865,
      "learning_rate": 9.494949494949495e-06,
      "loss": 2.5226,
      "mean_token_accuracy": 0.8662195056676865,
      "step": 406
    },
    {
      "epoch": 3.2560000000000002,
      "grad_norm": 6.032021999359131,
      "learning_rate": 9.393939393939394e-06,
      "loss": 2.4289,
      "mean_token_accuracy": 0.8762623816728592,
      "step": 407
    },
    {
      "epoch": 3.2640000000000002,
      "grad_norm": 6.2928948402404785,
      "learning_rate": 9.292929292929294e-06,
      "loss": 2.4797,
      "mean_token_accuracy": 0.8561984449625015,
      "step": 408
    },
    {
      "epoch": 3.2720000000000002,
      "grad_norm": 5.483267784118652,
      "learning_rate": 9.191919191919192e-06,
      "loss": 2.8567,
      "mean_token_accuracy": 0.8590980023145676,
      "step": 409
    },
    {
      "epoch": 3.2800000000000002,
      "grad_norm": 6.180997848510742,
      "learning_rate": 9.090909090909091e-06,
      "loss": 2.6063,
      "mean_token_accuracy": 0.8648645430803299,
      "step": 410
    },
    {
      "epoch": 3.288,
      "grad_norm": 6.12001371383667,
      "learning_rate": 8.98989898989899e-06,
      "loss": 2.6823,
      "mean_token_accuracy": 0.8513016700744629,
      "step": 411
    },
    {
      "epoch": 3.296,
      "grad_norm": 5.99603796005249,
      "learning_rate": 8.88888888888889e-06,
      "loss": 3.0047,
      "mean_token_accuracy": 0.8510095775127411,
      "step": 412
    },
    {
      "epoch": 3.304,
      "grad_norm": 5.749976634979248,
      "learning_rate": 8.787878787878788e-06,
      "loss": 2.9216,
      "mean_token_accuracy": 0.8540640473365784,
      "step": 413
    },
    {
      "epoch": 3.312,
      "grad_norm": 6.270049095153809,
      "learning_rate": 8.686868686868687e-06,
      "loss": 3.0965,
      "mean_token_accuracy": 0.8414539396762848,
      "step": 414
    },
    {
      "epoch": 3.32,
      "grad_norm": 5.532293319702148,
      "learning_rate": 8.585858585858587e-06,
      "loss": 3.0462,
      "mean_token_accuracy": 0.8508045822381973,
      "step": 415
    },
    {
      "epoch": 3.328,
      "grad_norm": 6.463992118835449,
      "learning_rate": 8.484848484848486e-06,
      "loss": 2.7394,
      "mean_token_accuracy": 0.85701984167099,
      "step": 416
    },
    {
      "epoch": 3.336,
      "grad_norm": 5.6121134757995605,
      "learning_rate": 8.383838383838384e-06,
      "loss": 2.7231,
      "mean_token_accuracy": 0.8531108349561691,
      "step": 417
    },
    {
      "epoch": 3.344,
      "grad_norm": 6.132601737976074,
      "learning_rate": 8.282828282828283e-06,
      "loss": 3.1036,
      "mean_token_accuracy": 0.8383132815361023,
      "step": 418
    },
    {
      "epoch": 3.352,
      "grad_norm": 6.347646713256836,
      "learning_rate": 8.181818181818183e-06,
      "loss": 2.5049,
      "mean_token_accuracy": 0.8660729080438614,
      "step": 419
    },
    {
      "epoch": 3.36,
      "grad_norm": 5.851092338562012,
      "learning_rate": 8.080808080808082e-06,
      "loss": 2.5479,
      "mean_token_accuracy": 0.8598389625549316,
      "step": 420
    },
    {
      "epoch": 3.368,
      "grad_norm": 5.76607084274292,
      "learning_rate": 7.97979797979798e-06,
      "loss": 3.1152,
      "mean_token_accuracy": 0.8485762178897858,
      "step": 421
    },
    {
      "epoch": 3.376,
      "grad_norm": 6.1592607498168945,
      "learning_rate": 7.878787878787878e-06,
      "loss": 2.514,
      "mean_token_accuracy": 0.8623416870832443,
      "step": 422
    },
    {
      "epoch": 3.384,
      "grad_norm": 6.193600654602051,
      "learning_rate": 7.777777777777777e-06,
      "loss": 2.9589,
      "mean_token_accuracy": 0.8406593501567841,
      "step": 423
    },
    {
      "epoch": 3.392,
      "grad_norm": 5.8620100021362305,
      "learning_rate": 7.676767676767677e-06,
      "loss": 2.6917,
      "mean_token_accuracy": 0.8631896525621414,
      "step": 424
    },
    {
      "epoch": 3.4,
      "grad_norm": 5.631361961364746,
      "learning_rate": 7.5757575757575764e-06,
      "loss": 2.6014,
      "mean_token_accuracy": 0.8725499212741852,
      "step": 425
    },
    {
      "epoch": 3.408,
      "grad_norm": 6.087384223937988,
      "learning_rate": 7.474747474747475e-06,
      "loss": 2.5906,
      "mean_token_accuracy": 0.8593233972787857,
      "step": 426
    },
    {
      "epoch": 3.416,
      "grad_norm": 6.2712016105651855,
      "learning_rate": 7.3737373737373745e-06,
      "loss": 2.7093,
      "mean_token_accuracy": 0.8598934262990952,
      "step": 427
    },
    {
      "epoch": 3.424,
      "grad_norm": 5.651885986328125,
      "learning_rate": 7.272727272727272e-06,
      "loss": 2.6062,
      "mean_token_accuracy": 0.8617094457149506,
      "step": 428
    },
    {
      "epoch": 3.432,
      "grad_norm": 5.506243705749512,
      "learning_rate": 7.171717171717173e-06,
      "loss": 2.5589,
      "mean_token_accuracy": 0.8667820692062378,
      "step": 429
    },
    {
      "epoch": 3.44,
      "grad_norm": 6.425995349884033,
      "learning_rate": 7.0707070707070704e-06,
      "loss": 3.1032,
      "mean_token_accuracy": 0.8365529775619507,
      "step": 430
    },
    {
      "epoch": 3.448,
      "grad_norm": 5.543388366699219,
      "learning_rate": 6.969696969696971e-06,
      "loss": 2.6756,
      "mean_token_accuracy": 0.8701014816761017,
      "step": 431
    },
    {
      "epoch": 3.456,
      "grad_norm": 6.339065074920654,
      "learning_rate": 6.8686868686868685e-06,
      "loss": 2.5654,
      "mean_token_accuracy": 0.8716783672571182,
      "step": 432
    },
    {
      "epoch": 3.464,
      "grad_norm": 6.019214153289795,
      "learning_rate": 6.767676767676769e-06,
      "loss": 3.3469,
      "mean_token_accuracy": 0.8329573720693588,
      "step": 433
    },
    {
      "epoch": 3.472,
      "grad_norm": 6.126465797424316,
      "learning_rate": 6.666666666666667e-06,
      "loss": 2.5695,
      "mean_token_accuracy": 0.8654420524835587,
      "step": 434
    },
    {
      "epoch": 3.48,
      "grad_norm": 5.879609107971191,
      "learning_rate": 6.565656565656567e-06,
      "loss": 2.5542,
      "mean_token_accuracy": 0.861924409866333,
      "step": 435
    },
    {
      "epoch": 3.488,
      "grad_norm": 6.507324695587158,
      "learning_rate": 6.464646464646465e-06,
      "loss": 2.8441,
      "mean_token_accuracy": 0.84620201587677,
      "step": 436
    },
    {
      "epoch": 3.496,
      "grad_norm": 6.121249675750732,
      "learning_rate": 6.363636363636363e-06,
      "loss": 2.2379,
      "mean_token_accuracy": 0.8786091357469559,
      "step": 437
    },
    {
      "epoch": 3.504,
      "grad_norm": 5.940850734710693,
      "learning_rate": 6.262626262626263e-06,
      "loss": 2.5193,
      "mean_token_accuracy": 0.8697729557752609,
      "step": 438
    },
    {
      "epoch": 3.512,
      "grad_norm": 6.618006229400635,
      "learning_rate": 6.161616161616162e-06,
      "loss": 3.0884,
      "mean_token_accuracy": 0.8440293669700623,
      "step": 439
    },
    {
      "epoch": 3.52,
      "grad_norm": 6.300545692443848,
      "learning_rate": 6.060606060606061e-06,
      "loss": 2.9795,
      "mean_token_accuracy": 0.8421255350112915,
      "step": 440
    },
    {
      "epoch": 3.528,
      "grad_norm": 6.084777355194092,
      "learning_rate": 5.9595959595959605e-06,
      "loss": 2.8443,
      "mean_token_accuracy": 0.8482642769813538,
      "step": 441
    },
    {
      "epoch": 3.536,
      "grad_norm": 6.6309685707092285,
      "learning_rate": 5.858585858585859e-06,
      "loss": 2.7261,
      "mean_token_accuracy": 0.8591820150613785,
      "step": 442
    },
    {
      "epoch": 3.544,
      "grad_norm": 5.60848331451416,
      "learning_rate": 5.7575757575757586e-06,
      "loss": 2.4302,
      "mean_token_accuracy": 0.8700147867202759,
      "step": 443
    },
    {
      "epoch": 3.552,
      "grad_norm": 6.067654132843018,
      "learning_rate": 5.656565656565657e-06,
      "loss": 2.5163,
      "mean_token_accuracy": 0.8643026798963547,
      "step": 444
    },
    {
      "epoch": 3.56,
      "grad_norm": 5.975160121917725,
      "learning_rate": 5.555555555555556e-06,
      "loss": 2.8837,
      "mean_token_accuracy": 0.8568368703126907,
      "step": 445
    },
    {
      "epoch": 3.568,
      "grad_norm": 6.040992736816406,
      "learning_rate": 5.4545454545454545e-06,
      "loss": 2.3984,
      "mean_token_accuracy": 0.867580771446228,
      "step": 446
    },
    {
      "epoch": 3.576,
      "grad_norm": 5.610777378082275,
      "learning_rate": 5.353535353535354e-06,
      "loss": 2.5171,
      "mean_token_accuracy": 0.8687434643507004,
      "step": 447
    },
    {
      "epoch": 3.584,
      "grad_norm": 6.74979829788208,
      "learning_rate": 5.2525252525252526e-06,
      "loss": 2.5815,
      "mean_token_accuracy": 0.866498276591301,
      "step": 448
    },
    {
      "epoch": 3.592,
      "grad_norm": 5.904205322265625,
      "learning_rate": 5.151515151515152e-06,
      "loss": 2.7649,
      "mean_token_accuracy": 0.8657421469688416,
      "step": 449
    },
    {
      "epoch": 3.6,
      "grad_norm": 5.820155620574951,
      "learning_rate": 5.050505050505051e-06,
      "loss": 2.7525,
      "mean_token_accuracy": 0.8619928807020187,
      "step": 450
    },
    {
      "epoch": 3.608,
      "grad_norm": 6.589330196380615,
      "learning_rate": 4.949494949494949e-06,
      "loss": 2.287,
      "mean_token_accuracy": 0.8719237148761749,
      "step": 451
    },
    {
      "epoch": 3.616,
      "grad_norm": 6.643290996551514,
      "learning_rate": 4.848484848484849e-06,
      "loss": 3.4465,
      "mean_token_accuracy": 0.841964989900589,
      "step": 452
    },
    {
      "epoch": 3.624,
      "grad_norm": 6.711119174957275,
      "learning_rate": 4.747474747474747e-06,
      "loss": 2.8413,
      "mean_token_accuracy": 0.856516107916832,
      "step": 453
    },
    {
      "epoch": 3.632,
      "grad_norm": 5.4684343338012695,
      "learning_rate": 4.646464646464647e-06,
      "loss": 2.4144,
      "mean_token_accuracy": 0.8840565979480743,
      "step": 454
    },
    {
      "epoch": 3.64,
      "grad_norm": 5.740659236907959,
      "learning_rate": 4.5454545454545455e-06,
      "loss": 2.6323,
      "mean_token_accuracy": 0.8655981123447418,
      "step": 455
    },
    {
      "epoch": 3.648,
      "grad_norm": 6.271495819091797,
      "learning_rate": 4.444444444444445e-06,
      "loss": 3.0839,
      "mean_token_accuracy": 0.8419709354639053,
      "step": 456
    },
    {
      "epoch": 3.656,
      "grad_norm": 5.493051052093506,
      "learning_rate": 4.343434343434344e-06,
      "loss": 2.1636,
      "mean_token_accuracy": 0.8905516117811203,
      "step": 457
    },
    {
      "epoch": 3.664,
      "grad_norm": 6.861065864562988,
      "learning_rate": 4.242424242424243e-06,
      "loss": 2.9378,
      "mean_token_accuracy": 0.8521191477775574,
      "step": 458
    },
    {
      "epoch": 3.672,
      "grad_norm": 6.059913158416748,
      "learning_rate": 4.141414141414142e-06,
      "loss": 2.7574,
      "mean_token_accuracy": 0.8573954999446869,
      "step": 459
    },
    {
      "epoch": 3.68,
      "grad_norm": 6.36575174331665,
      "learning_rate": 4.040404040404041e-06,
      "loss": 2.8077,
      "mean_token_accuracy": 0.8544747680425644,
      "step": 460
    },
    {
      "epoch": 3.6879999999999997,
      "grad_norm": 6.40785026550293,
      "learning_rate": 3.939393939393939e-06,
      "loss": 2.9,
      "mean_token_accuracy": 0.8473234623670578,
      "step": 461
    },
    {
      "epoch": 3.6959999999999997,
      "grad_norm": 6.139939785003662,
      "learning_rate": 3.8383838383838385e-06,
      "loss": 2.4719,
      "mean_token_accuracy": 0.8682427853345871,
      "step": 462
    },
    {
      "epoch": 3.7039999999999997,
      "grad_norm": 6.098149299621582,
      "learning_rate": 3.7373737373737375e-06,
      "loss": 2.7453,
      "mean_token_accuracy": 0.8645330965518951,
      "step": 463
    },
    {
      "epoch": 3.7119999999999997,
      "grad_norm": 6.3011884689331055,
      "learning_rate": 3.636363636363636e-06,
      "loss": 2.7872,
      "mean_token_accuracy": 0.8514862060546875,
      "step": 464
    },
    {
      "epoch": 3.7199999999999998,
      "grad_norm": 6.186530590057373,
      "learning_rate": 3.5353535353535352e-06,
      "loss": 3.1428,
      "mean_token_accuracy": 0.8448829352855682,
      "step": 465
    },
    {
      "epoch": 3.7279999999999998,
      "grad_norm": 7.000403881072998,
      "learning_rate": 3.4343434343434343e-06,
      "loss": 3.5395,
      "mean_token_accuracy": 0.8234449177980423,
      "step": 466
    },
    {
      "epoch": 3.7359999999999998,
      "grad_norm": 6.087681770324707,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 3.0334,
      "mean_token_accuracy": 0.8479331731796265,
      "step": 467
    },
    {
      "epoch": 3.7439999999999998,
      "grad_norm": 5.989120960235596,
      "learning_rate": 3.2323232323232324e-06,
      "loss": 2.7067,
      "mean_token_accuracy": 0.866983637213707,
      "step": 468
    },
    {
      "epoch": 3.752,
      "grad_norm": 6.254819869995117,
      "learning_rate": 3.1313131313131314e-06,
      "loss": 2.3856,
      "mean_token_accuracy": 0.8707826137542725,
      "step": 469
    },
    {
      "epoch": 3.76,
      "grad_norm": 6.099485874176025,
      "learning_rate": 3.0303030303030305e-06,
      "loss": 2.8972,
      "mean_token_accuracy": 0.8589896708726883,
      "step": 470
    },
    {
      "epoch": 3.768,
      "grad_norm": 6.424018383026123,
      "learning_rate": 2.9292929292929295e-06,
      "loss": 2.7964,
      "mean_token_accuracy": 0.8504993915557861,
      "step": 471
    },
    {
      "epoch": 3.776,
      "grad_norm": 5.645679950714111,
      "learning_rate": 2.8282828282828286e-06,
      "loss": 2.4383,
      "mean_token_accuracy": 0.8685364574193954,
      "step": 472
    },
    {
      "epoch": 3.784,
      "grad_norm": 6.7141523361206055,
      "learning_rate": 2.7272727272727272e-06,
      "loss": 2.8149,
      "mean_token_accuracy": 0.8539319187402725,
      "step": 473
    },
    {
      "epoch": 3.792,
      "grad_norm": 6.887962818145752,
      "learning_rate": 2.6262626262626263e-06,
      "loss": 2.6289,
      "mean_token_accuracy": 0.8702027946710587,
      "step": 474
    },
    {
      "epoch": 3.8,
      "grad_norm": 6.338537216186523,
      "learning_rate": 2.5252525252525253e-06,
      "loss": 3.0126,
      "mean_token_accuracy": 0.8451626300811768,
      "step": 475
    },
    {
      "epoch": 3.808,
      "grad_norm": 6.215012550354004,
      "learning_rate": 2.4242424242424244e-06,
      "loss": 3.0838,
      "mean_token_accuracy": 0.853987067937851,
      "step": 476
    },
    {
      "epoch": 3.816,
      "grad_norm": 6.2042999267578125,
      "learning_rate": 2.3232323232323234e-06,
      "loss": 2.962,
      "mean_token_accuracy": 0.8471638411283493,
      "step": 477
    },
    {
      "epoch": 3.824,
      "grad_norm": 5.406651020050049,
      "learning_rate": 2.2222222222222225e-06,
      "loss": 2.4559,
      "mean_token_accuracy": 0.872399777173996,
      "step": 478
    },
    {
      "epoch": 3.832,
      "grad_norm": 6.384584426879883,
      "learning_rate": 2.1212121212121216e-06,
      "loss": 2.7063,
      "mean_token_accuracy": 0.8628278374671936,
      "step": 479
    },
    {
      "epoch": 3.84,
      "grad_norm": 6.349820137023926,
      "learning_rate": 2.0202020202020206e-06,
      "loss": 2.3293,
      "mean_token_accuracy": 0.8819480836391449,
      "step": 480
    },
    {
      "epoch": 3.848,
      "grad_norm": 5.694690227508545,
      "learning_rate": 1.9191919191919192e-06,
      "loss": 2.554,
      "mean_token_accuracy": 0.8645013719797134,
      "step": 481
    },
    {
      "epoch": 3.856,
      "grad_norm": 6.480876445770264,
      "learning_rate": 1.818181818181818e-06,
      "loss": 3.0091,
      "mean_token_accuracy": 0.8551051169633865,
      "step": 482
    },
    {
      "epoch": 3.864,
      "grad_norm": 6.149241924285889,
      "learning_rate": 1.7171717171717171e-06,
      "loss": 3.0215,
      "mean_token_accuracy": 0.8434359133243561,
      "step": 483
    },
    {
      "epoch": 3.872,
      "grad_norm": 5.84627103805542,
      "learning_rate": 1.6161616161616162e-06,
      "loss": 2.7796,
      "mean_token_accuracy": 0.8535165041685104,
      "step": 484
    },
    {
      "epoch": 3.88,
      "grad_norm": 6.468245506286621,
      "learning_rate": 1.5151515151515152e-06,
      "loss": 2.8329,
      "mean_token_accuracy": 0.8529188930988312,
      "step": 485
    },
    {
      "epoch": 3.888,
      "grad_norm": 6.755955219268799,
      "learning_rate": 1.4141414141414143e-06,
      "loss": 3.3661,
      "mean_token_accuracy": 0.831203356385231,
      "step": 486
    },
    {
      "epoch": 3.896,
      "grad_norm": 6.618368148803711,
      "learning_rate": 1.3131313131313131e-06,
      "loss": 2.9949,
      "mean_token_accuracy": 0.8433407545089722,
      "step": 487
    },
    {
      "epoch": 3.904,
      "grad_norm": 6.448158264160156,
      "learning_rate": 1.2121212121212122e-06,
      "loss": 2.6009,
      "mean_token_accuracy": 0.8493270874023438,
      "step": 488
    },
    {
      "epoch": 3.912,
      "grad_norm": 6.494684219360352,
      "learning_rate": 1.1111111111111112e-06,
      "loss": 2.8263,
      "mean_token_accuracy": 0.847339928150177,
      "step": 489
    },
    {
      "epoch": 3.92,
      "grad_norm": 6.943580150604248,
      "learning_rate": 1.0101010101010103e-06,
      "loss": 3.0355,
      "mean_token_accuracy": 0.8504604995250702,
      "step": 490
    },
    {
      "epoch": 3.928,
      "grad_norm": 5.839914798736572,
      "learning_rate": 9.09090909090909e-07,
      "loss": 2.6543,
      "mean_token_accuracy": 0.8537915647029877,
      "step": 491
    },
    {
      "epoch": 3.936,
      "grad_norm": 5.800342082977295,
      "learning_rate": 8.080808080808081e-07,
      "loss": 2.4919,
      "mean_token_accuracy": 0.8684473484754562,
      "step": 492
    },
    {
      "epoch": 3.944,
      "grad_norm": 6.721318244934082,
      "learning_rate": 7.070707070707071e-07,
      "loss": 2.7316,
      "mean_token_accuracy": 0.8631936460733414,
      "step": 493
    },
    {
      "epoch": 3.952,
      "grad_norm": 5.557168960571289,
      "learning_rate": 6.060606060606061e-07,
      "loss": 2.7026,
      "mean_token_accuracy": 0.8652904033660889,
      "step": 494
    },
    {
      "epoch": 3.96,
      "grad_norm": 6.060300350189209,
      "learning_rate": 5.050505050505052e-07,
      "loss": 2.8709,
      "mean_token_accuracy": 0.854452446103096,
      "step": 495
    },
    {
      "epoch": 3.968,
      "grad_norm": 5.997809886932373,
      "learning_rate": 4.0404040404040405e-07,
      "loss": 2.5419,
      "mean_token_accuracy": 0.8659024238586426,
      "step": 496
    },
    {
      "epoch": 3.976,
      "grad_norm": 5.681960105895996,
      "learning_rate": 3.0303030303030305e-07,
      "loss": 2.7244,
      "mean_token_accuracy": 0.8663401901721954,
      "step": 497
    },
    {
      "epoch": 3.984,
      "grad_norm": 8.17302417755127,
      "learning_rate": 2.0202020202020202e-07,
      "loss": 3.1581,
      "mean_token_accuracy": 0.8333937376737595,
      "step": 498
    },
    {
      "epoch": 3.992,
      "grad_norm": 6.531320095062256,
      "learning_rate": 1.0101010101010101e-07,
      "loss": 3.0756,
      "mean_token_accuracy": 0.8531923592090607,
      "step": 499
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.187903881072998,
      "learning_rate": 0.0,
      "loss": 2.5067,
      "mean_token_accuracy": 0.8733052164316177,
      "step": 500
    }
  ],
  "logging_steps": 1,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 2610056134656000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}