{
  "best_metric": 0.6970161199569702,
  "best_model_checkpoint": "data/Llama-31-8B_task-2_180-samples_config-1/checkpoint-85",
  "epoch": 12.0,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.058823529411764705,
      "grad_norm": 0.7735174298286438,
      "learning_rate": 1.1764705882352942e-06,
      "loss": 1.1412,
      "step": 1
    },
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 0.6781718134880066,
      "learning_rate": 2.3529411764705885e-06,
      "loss": 1.1733,
      "step": 2
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 0.8626628518104553,
      "learning_rate": 4.705882352941177e-06,
      "loss": 1.166,
      "step": 4
    },
    {
      "epoch": 0.35294117647058826,
      "grad_norm": 0.7292124629020691,
      "learning_rate": 7.058823529411765e-06,
      "loss": 1.1735,
      "step": 6
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 0.7384842038154602,
      "learning_rate": 9.411764705882354e-06,
      "loss": 1.0963,
      "step": 8
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 0.6787011027336121,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 1.1215,
      "step": 10
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 0.6661468148231506,
      "learning_rate": 1.411764705882353e-05,
      "loss": 1.0333,
      "step": 12
    },
    {
      "epoch": 0.8235294117647058,
      "grad_norm": 0.6215376257896423,
      "learning_rate": 1.647058823529412e-05,
      "loss": 1.078,
      "step": 14
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 0.6646621227264404,
      "learning_rate": 1.8823529411764708e-05,
      "loss": 0.9476,
      "step": 16
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.0002387762069702,
      "eval_runtime": 37.0103,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 17
    },
    {
      "epoch": 1.0588235294117647,
      "grad_norm": 0.5615031719207764,
      "learning_rate": 2.1176470588235296e-05,
      "loss": 1.0218,
      "step": 18
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.4883713126182556,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 0.9748,
      "step": 20
    },
    {
      "epoch": 1.2941176470588236,
      "grad_norm": 0.3860337734222412,
      "learning_rate": 2.5882352941176475e-05,
      "loss": 0.941,
      "step": 22
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 0.34377098083496094,
      "learning_rate": 2.823529411764706e-05,
      "loss": 0.9078,
      "step": 24
    },
    {
      "epoch": 1.5294117647058822,
      "grad_norm": 0.3344781696796417,
      "learning_rate": 3.058823529411765e-05,
      "loss": 0.8314,
      "step": 26
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 0.3371179401874542,
      "learning_rate": 3.294117647058824e-05,
      "loss": 0.8386,
      "step": 28
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 0.3457156717777252,
      "learning_rate": 3.529411764705883e-05,
      "loss": 0.9142,
      "step": 30
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 0.3432137668132782,
      "learning_rate": 3.7647058823529415e-05,
      "loss": 0.789,
      "step": 32
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.37301379442214966,
      "learning_rate": 4e-05,
      "loss": 0.8652,
      "step": 34
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8278143405914307,
      "eval_runtime": 37.0111,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 34
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.34446483850479126,
      "learning_rate": 4.235294117647059e-05,
      "loss": 0.796,
      "step": 36
    },
    {
      "epoch": 2.235294117647059,
      "grad_norm": 0.29265981912612915,
      "learning_rate": 4.470588235294118e-05,
      "loss": 0.786,
      "step": 38
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.2872772216796875,
      "learning_rate": 4.705882352941177e-05,
      "loss": 0.7274,
      "step": 40
    },
    {
      "epoch": 2.4705882352941178,
      "grad_norm": 0.2969878017902374,
      "learning_rate": 4.9411764705882355e-05,
      "loss": 0.7753,
      "step": 42
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.3672356605529785,
      "learning_rate": 5.176470588235295e-05,
      "loss": 0.75,
      "step": 44
    },
    {
      "epoch": 2.7058823529411766,
      "grad_norm": 0.3807593882083893,
      "learning_rate": 5.411764705882353e-05,
      "loss": 0.7897,
      "step": 46
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.294073224067688,
      "learning_rate": 5.647058823529412e-05,
      "loss": 0.7733,
      "step": 48
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 0.35462719202041626,
      "learning_rate": 5.882352941176471e-05,
      "loss": 0.7407,
      "step": 50
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.750808596611023,
      "eval_runtime": 37.0098,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 51
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.29193738102912903,
      "learning_rate": 6.11764705882353e-05,
      "loss": 0.6996,
      "step": 52
    },
    {
      "epoch": 3.176470588235294,
      "grad_norm": 0.318910151720047,
      "learning_rate": 6.352941176470588e-05,
      "loss": 0.7119,
      "step": 54
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.306405633687973,
      "learning_rate": 6.588235294117648e-05,
      "loss": 0.6928,
      "step": 56
    },
    {
      "epoch": 3.411764705882353,
      "grad_norm": 0.39580562710762024,
      "learning_rate": 6.823529411764707e-05,
      "loss": 0.7475,
      "step": 58
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.4000554382801056,
      "learning_rate": 7.058823529411765e-05,
      "loss": 0.6553,
      "step": 60
    },
    {
      "epoch": 3.6470588235294117,
      "grad_norm": 0.3609679341316223,
      "learning_rate": 7.294117647058823e-05,
      "loss": 0.7201,
      "step": 62
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.4110010862350464,
      "learning_rate": 7.529411764705883e-05,
      "loss": 0.6693,
      "step": 64
    },
    {
      "epoch": 3.8823529411764706,
      "grad_norm": 0.38782674074172974,
      "learning_rate": 7.764705882352942e-05,
      "loss": 0.6769,
      "step": 66
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.39094892144203186,
      "learning_rate": 8e-05,
      "loss": 0.5985,
      "step": 68
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.710180938243866,
      "eval_runtime": 37.0027,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 68
    },
    {
      "epoch": 4.117647058823529,
      "grad_norm": 0.3916146457195282,
      "learning_rate": 8.23529411764706e-05,
      "loss": 0.6828,
      "step": 70
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.3907603919506073,
      "learning_rate": 8.470588235294118e-05,
      "loss": 0.5952,
      "step": 72
    },
    {
      "epoch": 4.352941176470588,
      "grad_norm": 0.3995077311992645,
      "learning_rate": 8.705882352941177e-05,
      "loss": 0.6093,
      "step": 74
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.4571545720100403,
      "learning_rate": 8.941176470588236e-05,
      "loss": 0.6,
      "step": 76
    },
    {
      "epoch": 4.588235294117647,
      "grad_norm": 0.472224622964859,
      "learning_rate": 9.176470588235295e-05,
      "loss": 0.5839,
      "step": 78
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.4620254933834076,
      "learning_rate": 9.411764705882353e-05,
      "loss": 0.5496,
      "step": 80
    },
    {
      "epoch": 4.823529411764706,
      "grad_norm": 0.5443108677864075,
      "learning_rate": 9.647058823529412e-05,
      "loss": 0.5749,
      "step": 82
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.5169690847396851,
      "learning_rate": 9.882352941176471e-05,
      "loss": 0.5758,
      "step": 84
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.6970161199569702,
      "eval_runtime": 37.0126,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 85
    },
    {
      "epoch": 5.0588235294117645,
      "grad_norm": 0.5234110951423645,
      "learning_rate": 9.99995783847866e-05,
      "loss": 0.5465,
      "step": 86
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.5332862734794617,
      "learning_rate": 9.999620550574153e-05,
      "loss": 0.4939,
      "step": 88
    },
    {
      "epoch": 5.294117647058823,
      "grad_norm": 0.5728622078895569,
      "learning_rate": 9.998945997517956e-05,
      "loss": 0.497,
      "step": 90
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.6428951025009155,
      "learning_rate": 9.997934224814173e-05,
      "loss": 0.5159,
      "step": 92
    },
    {
      "epoch": 5.529411764705882,
      "grad_norm": 0.5943310260772705,
      "learning_rate": 9.996585300715116e-05,
      "loss": 0.4513,
      "step": 94
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.6429039239883423,
      "learning_rate": 9.994899316216708e-05,
      "loss": 0.4521,
      "step": 96
    },
    {
      "epoch": 5.764705882352941,
      "grad_norm": 0.7048048973083496,
      "learning_rate": 9.992876385052345e-05,
      "loss": 0.4874,
      "step": 98
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.6601319909095764,
      "learning_rate": 9.990516643685222e-05,
      "loss": 0.496,
      "step": 100
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6444077491760254,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.4534,
      "step": 102
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.7430355548858643,
      "eval_runtime": 37.0203,
      "eval_samples_per_second": 0.972,
      "eval_steps_per_second": 0.972,
      "step": 102
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.6521551609039307,
      "learning_rate": 9.984787389787688e-05,
      "loss": 0.3221,
      "step": 104
    },
    {
      "epoch": 6.235294117647059,
      "grad_norm": 1.2075269222259521,
      "learning_rate": 9.981418263742148e-05,
      "loss": 0.32,
      "step": 106
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.8351272940635681,
      "learning_rate": 9.977713100437509e-05,
      "loss": 0.2945,
      "step": 108
    },
    {
      "epoch": 6.470588235294118,
      "grad_norm": 0.7331373691558838,
      "learning_rate": 9.973672149817232e-05,
      "loss": 0.3325,
      "step": 110
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.8742902278900146,
      "learning_rate": 9.96929568447637e-05,
      "loss": 0.3079,
      "step": 112
    },
    {
      "epoch": 6.705882352941177,
      "grad_norm": 0.8754763603210449,
      "learning_rate": 9.964583999643174e-05,
      "loss": 0.3469,
      "step": 114
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.9281278848648071,
      "learning_rate": 9.95953741315919e-05,
      "loss": 0.3537,
      "step": 116
    },
    {
      "epoch": 6.9411764705882355,
      "grad_norm": 0.8638790845870972,
      "learning_rate": 9.954156265457801e-05,
      "loss": 0.3294,
      "step": 118
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.7982418537139893,
      "eval_runtime": 37.0233,
      "eval_samples_per_second": 0.972,
      "eval_steps_per_second": 0.972,
      "step": 119
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.8910298943519592,
      "learning_rate": 9.948440919541278e-05,
      "loss": 0.2837,
      "step": 120
    },
    {
      "epoch": 7.176470588235294,
      "grad_norm": 0.8688742518424988,
      "learning_rate": 9.942391760956277e-05,
      "loss": 0.1808,
      "step": 122
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 1.6008689403533936,
      "learning_rate": 9.936009197767845e-05,
      "loss": 0.208,
      "step": 124
    },
    {
      "epoch": 7.411764705882353,
      "grad_norm": 1.4348896741867065,
      "learning_rate": 9.929293660531888e-05,
      "loss": 0.1646,
      "step": 126
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.9446174502372742,
      "learning_rate": 9.922245602266118e-05,
      "loss": 0.1597,
      "step": 128
    },
    {
      "epoch": 7.647058823529412,
      "grad_norm": 0.8924188613891602,
      "learning_rate": 9.91486549841951e-05,
      "loss": 0.1763,
      "step": 130
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.9412318468093872,
      "learning_rate": 9.90715384684021e-05,
      "loss": 0.1551,
      "step": 132
    },
    {
      "epoch": 7.882352941176471,
      "grad_norm": 1.3223024606704712,
      "learning_rate": 9.899111167741966e-05,
      "loss": 0.2041,
      "step": 134
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.1234177350997925,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.1775,
      "step": 136
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.0452101230621338,
      "eval_runtime": 37.0104,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 136
    },
    {
      "epoch": 8.117647058823529,
      "grad_norm": 1.0078459978103638,
      "learning_rate": 9.882034919459555e-05,
      "loss": 0.1009,
      "step": 138
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.9175763726234436,
      "learning_rate": 9.873002502207503e-05,
      "loss": 0.089,
      "step": 140
    },
    {
      "epoch": 8.352941176470589,
      "grad_norm": 1.2927151918411255,
      "learning_rate": 9.863641361223024e-05,
      "loss": 0.1013,
      "step": 142
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 1.0925740003585815,
      "learning_rate": 9.853952127991372e-05,
      "loss": 0.0991,
      "step": 144
    },
    {
      "epoch": 8.588235294117647,
      "grad_norm": 0.9681015014648438,
      "learning_rate": 9.843935456130295e-05,
      "loss": 0.0972,
      "step": 146
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.940649688243866,
      "learning_rate": 9.833592021345937e-05,
      "loss": 0.0807,
      "step": 148
    },
    {
      "epoch": 8.823529411764707,
      "grad_norm": 0.9116007089614868,
      "learning_rate": 9.822922521387276e-05,
      "loss": 0.0747,
      "step": 150
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 1.1829369068145752,
      "learning_rate": 9.811927675999036e-05,
      "loss": 0.0725,
      "step": 152
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.186506748199463,
      "eval_runtime": 37.0165,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 153
    },
    {
      "epoch": 9.058823529411764,
      "grad_norm": 0.6726549863815308,
      "learning_rate": 9.800608226873142e-05,
      "loss": 0.063,
      "step": 154
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.6709467172622681,
      "learning_rate": 9.788964937598689e-05,
      "loss": 0.0451,
      "step": 156
    },
    {
      "epoch": 9.294117647058824,
      "grad_norm": 1.0551162958145142,
      "learning_rate": 9.776998593610428e-05,
      "loss": 0.0395,
      "step": 158
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 1.0919300317764282,
      "learning_rate": 9.764710002135784e-05,
      "loss": 0.0411,
      "step": 160
    },
    {
      "epoch": 9.529411764705882,
      "grad_norm": 1.065580129623413,
      "learning_rate": 9.752099992140399e-05,
      "loss": 0.047,
      "step": 162
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.9504215121269226,
      "learning_rate": 9.739169414272217e-05,
      "loss": 0.0379,
      "step": 164
    },
    {
      "epoch": 9.764705882352942,
      "grad_norm": 0.7679919600486755,
      "learning_rate": 9.725919140804099e-05,
      "loss": 0.051,
      "step": 166
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.9968045949935913,
      "learning_rate": 9.71235006557497e-05,
      "loss": 0.0465,
      "step": 168
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.7186422348022461,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0479,
      "step": 170
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.2553839683532715,
      "eval_runtime": 37.019,
      "eval_samples_per_second": 0.972,
      "eval_steps_per_second": 0.972,
      "step": 170
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.6831572651863098,
      "learning_rate": 9.684259192656553e-05,
      "loss": 0.0256,
      "step": 172
    },
    {
      "epoch": 10.235294117647058,
      "grad_norm": 0.648961067199707,
      "learning_rate": 9.669739289925577e-05,
      "loss": 0.0245,
      "step": 174
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.6470946669578552,
      "learning_rate": 9.654904375222385e-05,
      "loss": 0.0263,
      "step": 176
    },
    {
      "epoch": 10.470588235294118,
      "grad_norm": 0.7383893132209778,
      "learning_rate": 9.639755449282875e-05,
      "loss": 0.0203,
      "step": 178
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 0.8391593098640442,
      "learning_rate": 9.62429353402556e-05,
      "loss": 0.0303,
      "step": 180
    },
    {
      "epoch": 10.705882352941176,
      "grad_norm": 0.7274612188339233,
      "learning_rate": 9.608519672482636e-05,
      "loss": 0.0207,
      "step": 182
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.8524788618087769,
      "learning_rate": 9.592434928729616e-05,
      "loss": 0.0289,
      "step": 184
    },
    {
      "epoch": 10.941176470588236,
      "grad_norm": 0.9105643630027771,
      "learning_rate": 9.576040387813552e-05,
      "loss": 0.033,
      "step": 186
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.4046423435211182,
      "eval_runtime": 37.0245,
      "eval_samples_per_second": 0.972,
      "eval_steps_per_second": 0.972,
      "step": 187
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.34708723425865173,
      "learning_rate": 9.559337155679842e-05,
      "loss": 0.0189,
      "step": 188
    },
    {
      "epoch": 11.176470588235293,
      "grad_norm": 0.6131543517112732,
      "learning_rate": 9.542326359097619e-05,
      "loss": 0.0168,
      "step": 190
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.7555575370788574,
      "learning_rate": 9.525009145583745e-05,
      "loss": 0.0194,
      "step": 192
    },
    {
      "epoch": 11.411764705882353,
      "grad_norm": 0.8476534485816956,
      "learning_rate": 9.507386683325404e-05,
      "loss": 0.0142,
      "step": 194
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.6849892735481262,
      "learning_rate": 9.489460161101291e-05,
      "loss": 0.0167,
      "step": 196
    },
    {
      "epoch": 11.647058823529411,
      "grad_norm": 0.9231868982315063,
      "learning_rate": 9.471230788201429e-05,
      "loss": 0.0157,
      "step": 198
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.7401719689369202,
      "learning_rate": 9.452699794345581e-05,
      "loss": 0.015,
      "step": 200
    },
    {
      "epoch": 11.882352941176471,
      "grad_norm": 0.836784303188324,
      "learning_rate": 9.43386842960031e-05,
      "loss": 0.0184,
      "step": 202
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.6760724186897278,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0153,
      "step": 204
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.513472080230713,
      "eval_runtime": 37.0204,
      "eval_samples_per_second": 0.972,
      "eval_steps_per_second": 0.972,
      "step": 204
    },
    {
      "epoch": 12.0,
      "step": 204,
      "total_flos": 1.562114679331881e+17,
      "train_loss": 0.4327789534193774,
      "train_runtime": 5577.7043,
      "train_samples_per_second": 1.219,
      "train_steps_per_second": 0.152
    }
  ],
  "logging_steps": 2,
  "max_steps": 850,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 25,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 7,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.562114679331881e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}