{
  "best_metric": 1.6509387493133545,
  "best_model_checkpoint": "/data/user_data/gonilude/oop_and_text_gpt2/checkpoint-250",
  "epoch": 3.0,
  "eval_steps": 50,
  "global_step": 717,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0,
      "eval_accuracy": 0.19811320754716982,
      "eval_loss": 8.76490592956543,
      "eval_runtime": 2.8023,
      "eval_samples_per_second": 75.653,
      "eval_steps_per_second": 9.635,
      "num_input_tokens_seen": 0,
      "step": 0
    },
    {
      "epoch": 0.0041841004184100415,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 6.0803,
      "num_input_tokens_seen": 8192,
      "step": 1
    },
    {
      "epoch": 0.02092050209205021,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 9.378,
      "num_input_tokens_seen": 40960,
      "step": 5
    },
    {
      "epoch": 0.04184100418410042,
      "grad_norm": 235.28628540039062,
      "learning_rate": 2.7272727272727272e-06,
      "loss": 5.6361,
      "num_input_tokens_seen": 81920,
      "step": 10
    },
    {
      "epoch": 0.06276150627615062,
      "grad_norm": 216.31581115722656,
      "learning_rate": 6.363636363636364e-06,
      "loss": 6.7539,
      "num_input_tokens_seen": 122880,
      "step": 15
    },
    {
      "epoch": 0.08368200836820083,
      "grad_norm": 283.3291015625,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 6.6226,
      "num_input_tokens_seen": 163840,
      "step": 20
    },
    {
      "epoch": 0.10460251046025104,
      "grad_norm": 181.8461151123047,
      "learning_rate": 1.5454545454545454e-05,
      "loss": 4.9251,
      "num_input_tokens_seen": 204800,
      "step": 25
    },
    {
      "epoch": 0.12552301255230125,
      "grad_norm": 87.72583770751953,
      "learning_rate": 2e-05,
      "loss": 3.5839,
      "num_input_tokens_seen": 245760,
      "step": 30
    },
    {
      "epoch": 0.14644351464435146,
      "grad_norm": 86.86190032958984,
      "learning_rate": 1.999744599547812e-05,
      "loss": 3.5158,
      "num_input_tokens_seen": 286720,
      "step": 35
    },
    {
      "epoch": 0.16736401673640167,
      "grad_norm": 43.51569366455078,
      "learning_rate": 1.9989785286500294e-05,
      "loss": 2.9512,
      "num_input_tokens_seen": 327680,
      "step": 40
    },
    {
      "epoch": 0.18828451882845187,
      "grad_norm": 57.58613967895508,
      "learning_rate": 1.99770217861636e-05,
      "loss": 2.2766,
      "num_input_tokens_seen": 368640,
      "step": 45
    },
    {
      "epoch": 0.20920502092050208,
      "grad_norm": 16.90093231201172,
      "learning_rate": 1.9959162014075553e-05,
      "loss": 1.6819,
      "num_input_tokens_seen": 409600,
      "step": 50
    },
    {
      "epoch": 0.20920502092050208,
      "eval_accuracy": 0.19339622641509435,
      "eval_loss": 3.081800937652588,
      "eval_runtime": 2.5932,
      "eval_samples_per_second": 81.753,
      "eval_steps_per_second": 10.412,
      "num_input_tokens_seen": 409600,
      "step": 50
    },
    {
      "epoch": 0.2301255230125523,
      "grad_norm": 13.574027061462402,
      "learning_rate": 1.9936215093023884e-05,
      "loss": 2.2788,
      "num_input_tokens_seen": 450560,
      "step": 55
    },
    {
      "epoch": 0.2510460251046025,
      "grad_norm": 15.552080154418945,
      "learning_rate": 1.990819274431662e-05,
      "loss": 2.2597,
      "num_input_tokens_seen": 491520,
      "step": 60
    },
    {
      "epoch": 0.2719665271966527,
      "grad_norm": 24.59514808654785,
      "learning_rate": 1.9875109281794828e-05,
      "loss": 1.8258,
      "num_input_tokens_seen": 532480,
      "step": 65
    },
    {
      "epoch": 0.2928870292887029,
      "grad_norm": 16.7853946685791,
      "learning_rate": 1.9836981604521077e-05,
      "loss": 2.1388,
      "num_input_tokens_seen": 573440,
      "step": 70
    },
    {
      "epoch": 0.3138075313807531,
      "grad_norm": 21.230756759643555,
      "learning_rate": 1.9793829188147406e-05,
      "loss": 1.9159,
      "num_input_tokens_seen": 614400,
      "step": 75
    },
    {
      "epoch": 0.33472803347280333,
      "grad_norm": 12.194311141967773,
      "learning_rate": 1.974567407496712e-05,
      "loss": 1.6805,
      "num_input_tokens_seen": 655360,
      "step": 80
    },
    {
      "epoch": 0.35564853556485354,
      "grad_norm": 11.538440704345703,
      "learning_rate": 1.9692540862655587e-05,
      "loss": 1.5511,
      "num_input_tokens_seen": 696320,
      "step": 85
    },
    {
      "epoch": 0.37656903765690375,
      "grad_norm": 9.909708976745605,
      "learning_rate": 1.9634456691705705e-05,
      "loss": 1.6952,
      "num_input_tokens_seen": 737280,
      "step": 90
    },
    {
      "epoch": 0.39748953974895396,
      "grad_norm": 14.122604370117188,
      "learning_rate": 1.9571451231564523e-05,
      "loss": 1.6081,
      "num_input_tokens_seen": 778240,
      "step": 95
    },
    {
      "epoch": 0.41841004184100417,
      "grad_norm": 12.841495513916016,
      "learning_rate": 1.9503556665478066e-05,
      "loss": 1.6936,
      "num_input_tokens_seen": 819200,
      "step": 100
    },
    {
      "epoch": 0.41841004184100417,
      "eval_accuracy": 0.2358490566037736,
      "eval_loss": 1.8988428115844727,
      "eval_runtime": 2.6014,
      "eval_samples_per_second": 81.494,
      "eval_steps_per_second": 10.379,
      "num_input_tokens_seen": 819200,
      "step": 100
    },
    {
      "epoch": 0.4393305439330544,
      "grad_norm": 13.155781745910645,
      "learning_rate": 1.9430807674052092e-05,
      "loss": 1.7461,
      "num_input_tokens_seen": 860160,
      "step": 105
    },
    {
      "epoch": 0.4602510460251046,
      "grad_norm": 15.09996509552002,
      "learning_rate": 1.9353241417537216e-05,
      "loss": 1.7519,
      "num_input_tokens_seen": 901120,
      "step": 110
    },
    {
      "epoch": 0.4811715481171548,
      "grad_norm": 11.305790901184082,
      "learning_rate": 1.9270897516847406e-05,
      "loss": 1.6317,
      "num_input_tokens_seen": 942080,
      "step": 115
    },
    {
      "epoch": 0.502092050209205,
      "grad_norm": 17.677629470825195,
      "learning_rate": 1.9183818033321612e-05,
      "loss": 1.8684,
      "num_input_tokens_seen": 983040,
      "step": 120
    },
    {
      "epoch": 0.5230125523012552,
      "grad_norm": 20.588193893432617,
      "learning_rate": 1.9092047447238775e-05,
      "loss": 1.8523,
      "num_input_tokens_seen": 1024000,
      "step": 125
    },
    {
      "epoch": 0.5439330543933054,
      "grad_norm": 8.660299301147461,
      "learning_rate": 1.899563263509725e-05,
      "loss": 1.6611,
      "num_input_tokens_seen": 1064960,
      "step": 130
    },
    {
      "epoch": 0.5648535564853556,
      "grad_norm": 13.271608352661133,
      "learning_rate": 1.8894622845670282e-05,
      "loss": 1.6906,
      "num_input_tokens_seen": 1105920,
      "step": 135
    },
    {
      "epoch": 0.5857740585774058,
      "grad_norm": 12.970966339111328,
      "learning_rate": 1.878906967484966e-05,
      "loss": 1.7494,
      "num_input_tokens_seen": 1146880,
      "step": 140
    },
    {
      "epoch": 0.606694560669456,
      "grad_norm": 12.701091766357422,
      "learning_rate": 1.86790270392905e-05,
      "loss": 1.745,
      "num_input_tokens_seen": 1187840,
      "step": 145
    },
    {
      "epoch": 0.6276150627615062,
      "grad_norm": 23.40630340576172,
      "learning_rate": 1.856455114887056e-05,
      "loss": 1.8252,
      "num_input_tokens_seen": 1228800,
      "step": 150
    },
    {
      "epoch": 0.6276150627615062,
      "eval_accuracy": 0.20754716981132076,
      "eval_loss": 1.7960366010665894,
      "eval_runtime": 2.5946,
      "eval_samples_per_second": 81.708,
      "eval_steps_per_second": 10.406,
      "num_input_tokens_seen": 1228800,
      "step": 150
    },
    {
      "epoch": 0.6485355648535565,
      "grad_norm": 15.101459503173828,
      "learning_rate": 1.8445700477978207e-05,
      "loss": 1.7943,
      "num_input_tokens_seen": 1269760,
      "step": 155
    },
    {
      "epoch": 0.6694560669456067,
      "grad_norm": 16.858856201171875,
      "learning_rate": 1.8322535735643604e-05,
      "loss": 1.7804,
      "num_input_tokens_seen": 1310720,
      "step": 160
    },
    {
      "epoch": 0.6903765690376569,
      "grad_norm": 8.641276359558105,
      "learning_rate": 1.8195119834528535e-05,
      "loss": 1.546,
      "num_input_tokens_seen": 1351680,
      "step": 165
    },
    {
      "epoch": 0.7112970711297071,
      "grad_norm": 21.16533660888672,
      "learning_rate": 1.8063517858790517e-05,
      "loss": 1.6679,
      "num_input_tokens_seen": 1392640,
      "step": 170
    },
    {
      "epoch": 0.7322175732217573,
      "grad_norm": 13.633495330810547,
      "learning_rate": 1.792779703083777e-05,
      "loss": 1.7481,
      "num_input_tokens_seen": 1433600,
      "step": 175
    },
    {
      "epoch": 0.7531380753138075,
      "grad_norm": 10.46967887878418,
      "learning_rate": 1.778802667699196e-05,
      "loss": 1.7758,
      "num_input_tokens_seen": 1474560,
      "step": 180
    },
    {
      "epoch": 0.7740585774058577,
      "grad_norm": 8.797440528869629,
      "learning_rate": 1.764427819207624e-05,
      "loss": 1.6425,
      "num_input_tokens_seen": 1515520,
      "step": 185
    },
    {
      "epoch": 0.7949790794979079,
      "grad_norm": 11.201558113098145,
      "learning_rate": 1.7496625002946702e-05,
      "loss": 1.6783,
      "num_input_tokens_seen": 1556480,
      "step": 190
    },
    {
      "epoch": 0.8158995815899581,
      "grad_norm": 14.996376991271973,
      "learning_rate": 1.734514253098589e-05,
      "loss": 1.6338,
      "num_input_tokens_seen": 1597440,
      "step": 195
    },
    {
      "epoch": 0.8368200836820083,
      "grad_norm": 11.189709663391113,
      "learning_rate": 1.7189908153577473e-05,
      "loss": 1.7185,
      "num_input_tokens_seen": 1638400,
      "step": 200
    },
    {
      "epoch": 0.8368200836820083,
      "eval_accuracy": 0.2169811320754717,
      "eval_loss": 1.6783148050308228,
      "eval_runtime": 2.6291,
      "eval_samples_per_second": 80.637,
      "eval_steps_per_second": 10.27,
      "num_input_tokens_seen": 1638400,
      "step": 200
    },
    {
      "epoch": 0.8577405857740585,
      "grad_norm": 7.823906898498535,
      "learning_rate": 1.7031001164581828e-05,
      "loss": 1.6345,
      "num_input_tokens_seen": 1679360,
      "step": 205
    },
    {
      "epoch": 0.8786610878661087,
      "grad_norm": 13.269351959228516,
      "learning_rate": 1.6868502733832647e-05,
      "loss": 1.6975,
      "num_input_tokens_seen": 1720320,
      "step": 210
    },
    {
      "epoch": 0.899581589958159,
      "grad_norm": 17.58881950378418,
      "learning_rate": 1.670249586567531e-05,
      "loss": 1.8369,
      "num_input_tokens_seen": 1761280,
      "step": 215
    },
    {
      "epoch": 0.9205020920502092,
      "grad_norm": 9.75393295288086,
      "learning_rate": 1.6533065356568206e-05,
      "loss": 1.5853,
      "num_input_tokens_seen": 1802240,
      "step": 220
    },
    {
      "epoch": 0.9414225941422594,
      "grad_norm": 12.639104843139648,
      "learning_rate": 1.636029775176862e-05,
      "loss": 1.6785,
      "num_input_tokens_seen": 1843200,
      "step": 225
    },
    {
      "epoch": 0.9623430962343096,
      "grad_norm": 15.948163986206055,
      "learning_rate": 1.618428130112533e-05,
      "loss": 1.6051,
      "num_input_tokens_seen": 1884160,
      "step": 230
    },
    {
      "epoch": 0.9832635983263598,
      "grad_norm": 8.31589126586914,
      "learning_rate": 1.6005105914000508e-05,
      "loss": 1.6842,
      "num_input_tokens_seen": 1925120,
      "step": 235
    },
    {
      "epoch": 1.00418410041841,
      "grad_norm": 19.15606689453125,
      "learning_rate": 1.5822863113343934e-05,
      "loss": 1.6257,
      "num_input_tokens_seen": 1966080,
      "step": 240
    },
    {
      "epoch": 1.0251046025104602,
      "grad_norm": 24.510881423950195,
      "learning_rate": 1.5637645988943008e-05,
      "loss": 1.7097,
      "num_input_tokens_seen": 2007040,
      "step": 245
    },
    {
      "epoch": 1.0460251046025104,
      "grad_norm": 11.306989669799805,
      "learning_rate": 1.544954914987238e-05,
      "loss": 1.6083,
      "num_input_tokens_seen": 2048000,
      "step": 250
    },
    {
      "epoch": 1.0460251046025104,
      "eval_accuracy": 0.16037735849056603,
      "eval_loss": 1.6509387493133545,
      "eval_runtime": 2.5472,
      "eval_samples_per_second": 83.227,
      "eval_steps_per_second": 10.6,
      "num_input_tokens_seen": 2048000,
      "step": 250
    },
    {
      "epoch": 1.0669456066945606,
      "grad_norm": 13.068288803100586,
      "learning_rate": 1.5258668676167548e-05,
      "loss": 1.5839,
      "num_input_tokens_seen": 2088960,
      "step": 255
    },
    {
      "epoch": 1.0878661087866108,
      "grad_norm": 13.336688995361328,
      "learning_rate": 1.5065102069747117e-05,
      "loss": 1.7474,
      "num_input_tokens_seen": 2129920,
      "step": 260
    },
    {
      "epoch": 1.108786610878661,
      "grad_norm": 21.65083122253418,
      "learning_rate": 1.48689482046087e-05,
      "loss": 1.5474,
      "num_input_tokens_seen": 2170880,
      "step": 265
    },
    {
      "epoch": 1.1297071129707112,
      "grad_norm": 11.06269359588623,
      "learning_rate": 1.467030727632401e-05,
      "loss": 1.5938,
      "num_input_tokens_seen": 2211840,
      "step": 270
    },
    {
      "epoch": 1.1506276150627615,
      "grad_norm": 4.230496406555176,
      "learning_rate": 1.4469280750858854e-05,
      "loss": 1.5958,
      "num_input_tokens_seen": 2252800,
      "step": 275
    },
    {
      "epoch": 1.1715481171548117,
      "grad_norm": 5.921239376068115,
      "learning_rate": 1.4265971312744252e-05,
      "loss": 1.6651,
      "num_input_tokens_seen": 2293760,
      "step": 280
    },
    {
      "epoch": 1.1924686192468619,
      "grad_norm": 16.350160598754883,
      "learning_rate": 1.4060482812625055e-05,
      "loss": 1.6057,
      "num_input_tokens_seen": 2334720,
      "step": 285
    },
    {
      "epoch": 1.213389121338912,
      "grad_norm": 12.474533081054688,
      "learning_rate": 1.3852920214212966e-05,
      "loss": 1.6118,
      "num_input_tokens_seen": 2375680,
      "step": 290
    },
    {
      "epoch": 1.2343096234309623,
      "grad_norm": 15.712911605834961,
      "learning_rate": 1.3643389540670963e-05,
      "loss": 1.5845,
      "num_input_tokens_seen": 2416640,
      "step": 295
    },
    {
      "epoch": 1.2552301255230125,
      "grad_norm": 8.5593900680542,
      "learning_rate": 1.3431997820456592e-05,
      "loss": 1.5989,
      "num_input_tokens_seen": 2457600,
      "step": 300
    },
    {
      "epoch": 1.2552301255230125,
      "eval_accuracy": 0.1792452830188679,
      "eval_loss": 1.7008217573165894,
      "eval_runtime": 2.5993,
      "eval_samples_per_second": 81.56,
      "eval_steps_per_second": 10.387,
      "num_input_tokens_seen": 2457600,
      "step": 300
    },
    {
      "epoch": 1.2761506276150627,
      "grad_norm": 9.534984588623047,
      "learning_rate": 1.3218853032651719e-05,
      "loss": 1.6117,
      "num_input_tokens_seen": 2498560,
      "step": 305
    },
    {
      "epoch": 1.297071129707113,
      "grad_norm": 15.189908027648926,
      "learning_rate": 1.3004064051806712e-05,
      "loss": 1.7553,
      "num_input_tokens_seen": 2539520,
      "step": 310
    },
    {
      "epoch": 1.3179916317991631,
      "grad_norm": 12.449243545532227,
      "learning_rate": 1.2787740592327232e-05,
      "loss": 1.665,
      "num_input_tokens_seen": 2580480,
      "step": 315
    },
    {
      "epoch": 1.3389121338912133,
      "grad_norm": 7.762625694274902,
      "learning_rate": 1.2569993152432028e-05,
      "loss": 1.5167,
      "num_input_tokens_seen": 2621440,
      "step": 320
    },
    {
      "epoch": 1.3598326359832635,
      "grad_norm": 16.984127044677734,
      "learning_rate": 1.2350932957710322e-05,
      "loss": 1.5832,
      "num_input_tokens_seen": 2662400,
      "step": 325
    },
    {
      "epoch": 1.3807531380753137,
      "grad_norm": 19.121389389038086,
      "learning_rate": 1.2130671904307692e-05,
      "loss": 1.6186,
      "num_input_tokens_seen": 2703360,
      "step": 330
    },
    {
      "epoch": 1.401673640167364,
      "grad_norm": 9.621400833129883,
      "learning_rate": 1.1909322501769407e-05,
      "loss": 1.6627,
      "num_input_tokens_seen": 2744320,
      "step": 335
    },
    {
      "epoch": 1.4225941422594142,
      "grad_norm": 10.77039909362793,
      "learning_rate": 1.1686997815570473e-05,
      "loss": 1.5581,
      "num_input_tokens_seen": 2785280,
      "step": 340
    },
    {
      "epoch": 1.4435146443514644,
      "grad_norm": 14.407941818237305,
      "learning_rate": 1.1463811409361667e-05,
      "loss": 1.6307,
      "num_input_tokens_seen": 2826240,
      "step": 345
    },
    {
      "epoch": 1.4644351464435146,
      "grad_norm": 20.046255111694336,
      "learning_rate": 1.1239877286961123e-05,
      "loss": 1.7146,
      "num_input_tokens_seen": 2867200,
      "step": 350
    },
    {
      "epoch": 1.4644351464435146,
      "eval_accuracy": 0.1792452830188679,
      "eval_loss": 1.7357338666915894,
      "eval_runtime": 2.5964,
      "eval_samples_per_second": 81.651,
      "eval_steps_per_second": 10.399,
      "num_input_tokens_seen": 2867200,
      "step": 350
    },
    {
      "epoch": 1.4853556485355648,
      "grad_norm": 7.564424991607666,
      "learning_rate": 1.1015309834121083e-05,
      "loss": 1.6814,
      "num_input_tokens_seen": 2908160,
      "step": 355
    },
    {
      "epoch": 1.506276150627615,
      "grad_norm": 14.892422676086426,
      "learning_rate": 1.079022376009955e-05,
      "loss": 1.6091,
      "num_input_tokens_seen": 2949120,
      "step": 360
    },
    {
      "epoch": 1.5271966527196654,
      "grad_norm": 13.213648796081543,
      "learning_rate": 1.05647340390667e-05,
      "loss": 1.6221,
      "num_input_tokens_seen": 2990080,
      "step": 365
    },
    {
      "epoch": 1.5481171548117154,
      "grad_norm": 13.863093376159668,
      "learning_rate": 1.0338955851375962e-05,
      "loss": 1.597,
      "num_input_tokens_seen": 3031040,
      "step": 370
    },
    {
      "epoch": 1.5690376569037658,
      "grad_norm": 10.547773361206055,
      "learning_rate": 1.01130045247298e-05,
      "loss": 1.6803,
      "num_input_tokens_seen": 3072000,
      "step": 375
    },
    {
      "epoch": 1.5899581589958158,
      "grad_norm": 12.884955406188965,
      "learning_rate": 9.886995475270205e-06,
      "loss": 1.5439,
      "num_input_tokens_seen": 3112960,
      "step": 380
    },
    {
      "epoch": 1.6108786610878663,
      "grad_norm": 19.711307525634766,
      "learning_rate": 9.661044148624038e-06,
      "loss": 1.8286,
      "num_input_tokens_seen": 3153920,
      "step": 385
    },
    {
      "epoch": 1.6317991631799162,
      "grad_norm": 16.134384155273438,
      "learning_rate": 9.435265960933304e-06,
      "loss": 1.7434,
      "num_input_tokens_seen": 3194880,
      "step": 390
    },
    {
      "epoch": 1.6527196652719667,
      "grad_norm": 6.306771755218506,
      "learning_rate": 9.209776239900453e-06,
      "loss": 1.6327,
      "num_input_tokens_seen": 3235840,
      "step": 395
    },
    {
      "epoch": 1.6736401673640167,
      "grad_norm": 11.366308212280273,
      "learning_rate": 8.98469016587892e-06,
      "loss": 1.5851,
      "num_input_tokens_seen": 3276800,
      "step": 400
    },
    {
      "epoch": 1.6736401673640167,
      "eval_accuracy": 0.20754716981132076,
      "eval_loss": 1.714836835861206,
      "eval_runtime": 2.6019,
      "eval_samples_per_second": 81.478,
      "eval_steps_per_second": 10.377,
      "num_input_tokens_seen": 3276800,
      "step": 400
    },
    {
      "epoch": 1.694560669456067,
      "grad_norm": 7.767678260803223,
      "learning_rate": 8.76012271303888e-06,
      "loss": 1.6208,
      "num_input_tokens_seen": 3317760,
      "step": 405
    },
    {
      "epoch": 1.715481171548117,
      "grad_norm": 12.073616981506348,
      "learning_rate": 8.536188590638334e-06,
      "loss": 1.6727,
      "num_input_tokens_seen": 3358720,
      "step": 410
    },
    {
      "epoch": 1.7364016736401675,
      "grad_norm": 16.02838897705078,
      "learning_rate": 8.313002184429529e-06,
      "loss": 1.6529,
      "num_input_tokens_seen": 3399680,
      "step": 415
    },
    {
      "epoch": 1.7573221757322175,
      "grad_norm": 13.083491325378418,
      "learning_rate": 8.090677498230598e-06,
      "loss": 1.6403,
      "num_input_tokens_seen": 3440640,
      "step": 420
    },
    {
      "epoch": 1.778242677824268,
      "grad_norm": 15.727888107299805,
      "learning_rate": 7.869328095692313e-06,
      "loss": 1.6371,
      "num_input_tokens_seen": 3481600,
      "step": 425
    },
    {
      "epoch": 1.799163179916318,
      "grad_norm": 11.377220153808594,
      "learning_rate": 7.649067042289681e-06,
      "loss": 1.7045,
      "num_input_tokens_seen": 3522560,
      "step": 430
    },
    {
      "epoch": 1.8200836820083683,
      "grad_norm": 10.347882270812988,
      "learning_rate": 7.430006847567972e-06,
      "loss": 1.6273,
      "num_input_tokens_seen": 3563520,
      "step": 435
    },
    {
      "epoch": 1.8410041841004183,
      "grad_norm": 11.076630592346191,
      "learning_rate": 7.2122594076727705e-06,
      "loss": 1.6002,
      "num_input_tokens_seen": 3604480,
      "step": 440
    },
    {
      "epoch": 1.8619246861924688,
      "grad_norm": 13.331921577453613,
      "learning_rate": 6.995935948193294e-06,
      "loss": 1.6046,
      "num_input_tokens_seen": 3645440,
      "step": 445
    },
    {
      "epoch": 1.8828451882845187,
      "grad_norm": 14.614879608154297,
      "learning_rate": 6.781146967348283e-06,
      "loss": 1.5185,
      "num_input_tokens_seen": 3686400,
      "step": 450
    },
    {
      "epoch": 1.8828451882845187,
      "eval_accuracy": 0.21226415094339623,
      "eval_loss": 1.663746953010559,
      "eval_runtime": 2.6002,
      "eval_samples_per_second": 81.531,
      "eval_steps_per_second": 10.384,
      "num_input_tokens_seen": 3686400,
      "step": 450
    },
    {
      "epoch": 1.9037656903765692,
      "grad_norm": 10.488274574279785,
      "learning_rate": 6.568002179543409e-06,
      "loss": 1.6239,
      "num_input_tokens_seen": 3727360,
      "step": 455
    },
    {
      "epoch": 1.9246861924686192,
      "grad_norm": 16.766536712646484,
      "learning_rate": 6.356610459329038e-06,
      "loss": 1.6541,
      "num_input_tokens_seen": 3768320,
      "step": 460
    },
    {
      "epoch": 1.9456066945606696,
      "grad_norm": 11.10879898071289,
      "learning_rate": 6.147079785787038e-06,
      "loss": 1.6157,
      "num_input_tokens_seen": 3809280,
      "step": 465
    },
    {
      "epoch": 1.9665271966527196,
      "grad_norm": 5.657884120941162,
      "learning_rate": 5.93951718737495e-06,
      "loss": 1.5612,
      "num_input_tokens_seen": 3850240,
      "step": 470
    },
    {
      "epoch": 1.98744769874477,
      "grad_norm": 16.088945388793945,
      "learning_rate": 5.7340286872557515e-06,
      "loss": 1.7009,
      "num_input_tokens_seen": 3891200,
      "step": 475
    },
    {
      "epoch": 2.00836820083682,
      "grad_norm": 9.804430961608887,
      "learning_rate": 5.530719249141148e-06,
      "loss": 1.7034,
      "num_input_tokens_seen": 3932160,
      "step": 480
    },
    {
      "epoch": 2.0292887029288704,
      "grad_norm": 12.8722505569458,
      "learning_rate": 5.329692723675994e-06,
      "loss": 1.6172,
      "num_input_tokens_seen": 3973120,
      "step": 485
    },
    {
      "epoch": 2.0502092050209204,
      "grad_norm": 14.757617950439453,
      "learning_rate": 5.131051795391302e-06,
      "loss": 1.5906,
      "num_input_tokens_seen": 4014080,
      "step": 490
    },
    {
      "epoch": 2.071129707112971,
      "grad_norm": 5.688968181610107,
      "learning_rate": 4.934897930252887e-06,
      "loss": 1.6135,
      "num_input_tokens_seen": 4055040,
      "step": 495
    },
    {
      "epoch": 2.092050209205021,
      "grad_norm": 17.570228576660156,
      "learning_rate": 4.7413313238324556e-06,
      "loss": 1.5803,
      "num_input_tokens_seen": 4096000,
      "step": 500
    },
    {
      "epoch": 2.092050209205021,
      "eval_accuracy": 0.18396226415094338,
      "eval_loss": 1.6518347263336182,
      "eval_runtime": 2.6033,
      "eval_samples_per_second": 81.434,
      "eval_steps_per_second": 10.371,
      "num_input_tokens_seen": 4096000,
      "step": 500
    },
    {
      "epoch": 2.1129707112970713,
      "grad_norm": 13.995955467224121,
      "learning_rate": 4.550450850127626e-06,
      "loss": 1.5606,
      "num_input_tokens_seen": 4136960,
      "step": 505
    },
    {
      "epoch": 2.1338912133891212,
      "grad_norm": 13.139900207519531,
      "learning_rate": 4.3623540110569935e-06,
      "loss": 1.5754,
      "num_input_tokens_seen": 4177920,
      "step": 510
    },
    {
      "epoch": 2.1548117154811717,
      "grad_norm": 13.75228214263916,
      "learning_rate": 4.177136886656067e-06,
      "loss": 1.6154,
      "num_input_tokens_seen": 4218880,
      "step": 515
    },
    {
      "epoch": 2.1757322175732217,
      "grad_norm": 9.623847007751465,
      "learning_rate": 3.9948940859994964e-06,
      "loss": 1.7061,
      "num_input_tokens_seen": 4259840,
      "step": 520
    },
    {
      "epoch": 2.196652719665272,
      "grad_norm": 11.617643356323242,
      "learning_rate": 3.815718698874672e-06,
      "loss": 1.6587,
      "num_input_tokens_seen": 4300800,
      "step": 525
    },
    {
      "epoch": 2.217573221757322,
      "grad_norm": 15.032342910766602,
      "learning_rate": 3.6397022482313804e-06,
      "loss": 1.6144,
      "num_input_tokens_seen": 4341760,
      "step": 530
    },
    {
      "epoch": 2.2384937238493725,
      "grad_norm": 9.221100807189941,
      "learning_rate": 3.466934643431795e-06,
      "loss": 1.57,
      "num_input_tokens_seen": 4382720,
      "step": 535
    },
    {
      "epoch": 2.2594142259414225,
      "grad_norm": 8.32654857635498,
      "learning_rate": 3.2975041343246937e-06,
      "loss": 1.6029,
      "num_input_tokens_seen": 4423680,
      "step": 540
    },
    {
      "epoch": 2.280334728033473,
      "grad_norm": 14.704009056091309,
      "learning_rate": 3.1314972661673572e-06,
      "loss": 1.5517,
      "num_input_tokens_seen": 4464640,
      "step": 545
    },
    {
      "epoch": 2.301255230125523,
      "grad_norm": 12.633601188659668,
      "learning_rate": 2.9689988354181742e-06,
      "loss": 1.5545,
      "num_input_tokens_seen": 4505600,
      "step": 550
    },
    {
      "epoch": 2.301255230125523,
      "eval_accuracy": 0.20754716981132076,
      "eval_loss": 1.6629799604415894,
      "eval_runtime": 2.6067,
      "eval_samples_per_second": 81.328,
      "eval_steps_per_second": 10.358,
      "num_input_tokens_seen": 4505600,
      "step": 550
    },
    {
      "epoch": 2.3221757322175733,
      "grad_norm": 13.853372573852539,
      "learning_rate": 2.8100918464225304e-06,
      "loss": 1.5857,
      "num_input_tokens_seen": 4546560,
      "step": 555
    },
    {
      "epoch": 2.3430962343096233,
      "grad_norm": 5.996284484863281,
      "learning_rate": 2.654857469014113e-06,
      "loss": 1.575,
      "num_input_tokens_seen": 4587520,
      "step": 560
    },
    {
      "epoch": 2.3640167364016738,
      "grad_norm": 14.616155624389648,
      "learning_rate": 2.5033749970533015e-06,
      "loss": 1.5556,
      "num_input_tokens_seen": 4628480,
      "step": 565
    },
    {
      "epoch": 2.3849372384937237,
      "grad_norm": 11.603802680969238,
      "learning_rate": 2.3557218079237608e-06,
      "loss": 1.6248,
      "num_input_tokens_seen": 4669440,
      "step": 570
    },
    {
      "epoch": 2.405857740585774,
      "grad_norm": 11.493803024291992,
      "learning_rate": 2.211973323008041e-06,
      "loss": 1.565,
      "num_input_tokens_seen": 4710400,
      "step": 575
    },
    {
      "epoch": 2.426778242677824,
      "grad_norm": 10.102631568908691,
      "learning_rate": 2.072202969162234e-06,
      "loss": 1.6141,
      "num_input_tokens_seen": 4751360,
      "step": 580
    },
    {
      "epoch": 2.4476987447698746,
      "grad_norm": 12.281519889831543,
      "learning_rate": 1.936482141209486e-06,
      "loss": 1.6051,
      "num_input_tokens_seen": 4792320,
      "step": 585
    },
    {
      "epoch": 2.4686192468619246,
      "grad_norm": 16.094688415527344,
      "learning_rate": 1.8048801654714687e-06,
      "loss": 1.5604,
      "num_input_tokens_seen": 4833280,
      "step": 590
    },
    {
      "epoch": 2.489539748953975,
      "grad_norm": 12.997512817382812,
      "learning_rate": 1.6774642643563955e-06,
      "loss": 1.5749,
      "num_input_tokens_seen": 4874240,
      "step": 595
    },
    {
      "epoch": 2.510460251046025,
      "grad_norm": 6.9834675788879395,
      "learning_rate": 1.5542995220217961e-06,
      "loss": 1.6077,
      "num_input_tokens_seen": 4915200,
      "step": 600
    },
    {
      "epoch": 2.510460251046025,
      "eval_accuracy": 0.18396226415094338,
      "eval_loss": 1.6762280464172363,
      "eval_runtime": 2.6046,
      "eval_samples_per_second": 81.395,
      "eval_steps_per_second": 10.366,
      "num_input_tokens_seen": 4915200,
      "step": 600
    },
    {
      "epoch": 2.5313807531380754,
      "grad_norm": 9.369280815124512,
      "learning_rate": 1.4354488511294418e-06,
      "loss": 1.5969,
      "num_input_tokens_seen": 4956160,
      "step": 605
    },
    {
      "epoch": 2.5523012552301254,
      "grad_norm": 14.795963287353516,
      "learning_rate": 1.3209729607095022e-06,
      "loss": 1.5874,
      "num_input_tokens_seen": 4997120,
      "step": 610
    },
    {
      "epoch": 2.573221757322176,
      "grad_norm": 9.430100440979004,
      "learning_rate": 1.2109303251503434e-06,
      "loss": 1.583,
      "num_input_tokens_seen": 5038080,
      "step": 615
    },
    {
      "epoch": 2.594142259414226,
      "grad_norm": 18.15077018737793,
      "learning_rate": 1.1053771543297198e-06,
      "loss": 1.5718,
      "num_input_tokens_seen": 5079040,
      "step": 620
    },
    {
      "epoch": 2.6150627615062763,
      "grad_norm": 17.04999542236328,
      "learning_rate": 1.0043673649027519e-06,
      "loss": 1.5646,
      "num_input_tokens_seen": 5120000,
      "step": 625
    },
    {
      "epoch": 2.6359832635983262,
      "grad_norm": 17.693492889404297,
      "learning_rate": 9.079525527612321e-07,
      "loss": 1.622,
      "num_input_tokens_seen": 5160960,
      "step": 630
    },
    {
      "epoch": 2.6569037656903767,
      "grad_norm": 13.240562438964844,
      "learning_rate": 8.161819666783888e-07,
      "loss": 1.6245,
      "num_input_tokens_seen": 5201920,
      "step": 635
    },
    {
      "epoch": 2.6778242677824267,
      "grad_norm": 6.269865989685059,
      "learning_rate": 7.291024831525961e-07,
      "loss": 1.5854,
      "num_input_tokens_seen": 5242880,
      "step": 640
    },
    {
      "epoch": 2.698744769874477,
      "grad_norm": 11.940990447998047,
      "learning_rate": 6.467585824627886e-07,
      "loss": 1.5721,
      "num_input_tokens_seen": 5283840,
      "step": 645
    },
    {
      "epoch": 2.719665271966527,
      "grad_norm": 9.96033763885498,
      "learning_rate": 5.691923259479093e-07,
      "loss": 1.6173,
      "num_input_tokens_seen": 5324800,
      "step": 650
    },
    {
      "epoch": 2.719665271966527,
      "eval_accuracy": 0.19339622641509435,
      "eval_loss": 1.674668788909912,
      "eval_runtime": 2.6008,
      "eval_samples_per_second": 81.512,
      "eval_steps_per_second": 10.381,
      "num_input_tokens_seen": 5324800,
      "step": 650
    },
    {
      "epoch": 2.7405857740585775,
      "grad_norm": 13.518848419189453,
      "learning_rate": 4.964433345219354e-07,
      "loss": 1.5959,
      "num_input_tokens_seen": 5365760,
      "step": 655
    },
    {
      "epoch": 2.7615062761506275,
      "grad_norm": 11.256808280944824,
      "learning_rate": 4.285487684354772e-07,
      "loss": 1.5775,
      "num_input_tokens_seen": 5406720,
      "step": 660
    },
    {
      "epoch": 2.782426778242678,
      "grad_norm": 13.231344223022461,
      "learning_rate": 3.6554330829429716e-07,
      "loss": 1.612,
      "num_input_tokens_seen": 5447680,
      "step": 665
    },
    {
      "epoch": 2.803347280334728,
      "grad_norm": 10.038106918334961,
      "learning_rate": 3.0745913734441357e-07,
      "loss": 1.5344,
      "num_input_tokens_seen": 5488640,
      "step": 670
    },
    {
      "epoch": 2.8242677824267783,
      "grad_norm": 6.0467963218688965,
      "learning_rate": 2.5432592503288e-07,
      "loss": 1.5885,
      "num_input_tokens_seen": 5529600,
      "step": 675
    },
    {
      "epoch": 2.8451882845188283,
      "grad_norm": 10.16049575805664,
      "learning_rate": 2.0617081185259512e-07,
      "loss": 1.5695,
      "num_input_tokens_seen": 5570560,
      "step": 680
    },
    {
      "epoch": 2.8661087866108788,
      "grad_norm": 17.670150756835938,
      "learning_rate": 1.630183954789233e-07,
      "loss": 1.6075,
      "num_input_tokens_seen": 5611520,
      "step": 685
    },
    {
      "epoch": 2.8870292887029287,
      "grad_norm": 16.916080474853516,
      "learning_rate": 1.2489071820517394e-07,
      "loss": 1.5717,
      "num_input_tokens_seen": 5652480,
      "step": 690
    },
    {
      "epoch": 2.907949790794979,
      "grad_norm": 10.792778015136719,
      "learning_rate": 9.180725568338045e-08,
      "loss": 1.5042,
      "num_input_tokens_seen": 5693440,
      "step": 695
    },
    {
      "epoch": 2.928870292887029,
      "grad_norm": 13.165460586547852,
      "learning_rate": 6.378490697611761e-08,
      "loss": 1.5865,
      "num_input_tokens_seen": 5734400,
      "step": 700
    },
    {
      "epoch": 2.928870292887029,
      "eval_accuracy": 0.19811320754716982,
      "eval_loss": 1.6730035543441772,
      "eval_runtime": 2.6248,
      "eval_samples_per_second": 80.768,
      "eval_steps_per_second": 10.286,
      "num_input_tokens_seen": 5734400,
      "step": 700
    },
    {
      "epoch": 2.9497907949790796,
      "grad_norm": 11.033462524414062,
      "learning_rate": 4.083798592444899e-08,
      "loss": 1.5476,
      "num_input_tokens_seen": 5775360,
      "step": 705
    },
    {
      "epoch": 2.9707112970711296,
      "grad_norm": 9.883075714111328,
      "learning_rate": 2.2978213836400974e-08,
      "loss": 1.5777,
      "num_input_tokens_seen": 5816320,
      "step": 710
    },
    {
      "epoch": 2.99163179916318,
      "grad_norm": 14.289135932922363,
      "learning_rate": 1.0214713499706596e-08,
      "loss": 1.6006,
      "num_input_tokens_seen": 5857280,
      "step": 715
    },
    {
      "epoch": 3.0,
      "num_input_tokens_seen": 5873664,
      "step": 717,
      "total_flos": 1.0654234806583296e+16,
      "train_loss": 1.861144894502818,
      "train_runtime": 604.4348,
      "train_samples_per_second": 9.455,
      "train_steps_per_second": 1.186
    }
  ],
  "logging_steps": 5,
  "max_steps": 717,
  "num_input_tokens_seen": 5873664,
  "num_train_epochs": 3,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0654234806583296e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}