{
  "best_global_step": 456,
  "best_metric": 0.24273568391799927,
  "best_model_checkpoint": "saves/prefix-tuning/llama-3-8b-instruct/train_cb_1756729608/checkpoint-456",
  "epoch": 10.0,
  "eval_steps": 57,
  "global_step": 1130,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04424778761061947,
      "grad_norm": 123.47064208984375,
      "learning_rate": 1.7699115044247788e-06,
      "loss": 9.9115,
      "num_input_tokens_seen": 1296,
      "step": 5
    },
    {
      "epoch": 0.08849557522123894,
      "grad_norm": 98.3427505493164,
      "learning_rate": 3.982300884955752e-06,
      "loss": 8.7072,
      "num_input_tokens_seen": 2944,
      "step": 10
    },
    {
      "epoch": 0.13274336283185842,
      "grad_norm": 77.04097747802734,
      "learning_rate": 6.194690265486726e-06,
      "loss": 6.9914,
      "num_input_tokens_seen": 4544,
      "step": 15
    },
    {
      "epoch": 0.17699115044247787,
      "grad_norm": 71.6903076171875,
      "learning_rate": 8.407079646017701e-06,
      "loss": 4.7551,
      "num_input_tokens_seen": 6048,
      "step": 20
    },
    {
      "epoch": 0.22123893805309736,
      "grad_norm": 73.29718780517578,
      "learning_rate": 1.0619469026548673e-05,
      "loss": 2.6531,
      "num_input_tokens_seen": 7296,
      "step": 25
    },
    {
      "epoch": 0.26548672566371684,
      "grad_norm": 60.493263244628906,
      "learning_rate": 1.2831858407079647e-05,
      "loss": 1.1721,
      "num_input_tokens_seen": 8944,
      "step": 30
    },
    {
      "epoch": 0.30973451327433627,
      "grad_norm": 16.882484436035156,
      "learning_rate": 1.504424778761062e-05,
      "loss": 0.3317,
      "num_input_tokens_seen": 10848,
      "step": 35
    },
    {
      "epoch": 0.35398230088495575,
      "grad_norm": 62.05000686645508,
      "learning_rate": 1.7256637168141594e-05,
      "loss": 0.387,
      "num_input_tokens_seen": 12416,
      "step": 40
    },
    {
      "epoch": 0.39823008849557523,
      "grad_norm": 51.86784744262695,
      "learning_rate": 1.946902654867257e-05,
      "loss": 0.2767,
      "num_input_tokens_seen": 14064,
      "step": 45
    },
    {
      "epoch": 0.4424778761061947,
      "grad_norm": 6.950078010559082,
      "learning_rate": 2.1681415929203542e-05,
      "loss": 1.0701,
      "num_input_tokens_seen": 15504,
      "step": 50
    },
    {
      "epoch": 0.48672566371681414,
      "grad_norm": 37.71853256225586,
      "learning_rate": 2.3893805309734516e-05,
      "loss": 0.6373,
      "num_input_tokens_seen": 16608,
      "step": 55
    },
    {
      "epoch": 0.504424778761062,
      "eval_loss": 0.4306437373161316,
      "eval_runtime": 0.7538,
      "eval_samples_per_second": 33.167,
      "eval_steps_per_second": 17.247,
      "num_input_tokens_seen": 17136,
      "step": 57
    },
    {
      "epoch": 0.5309734513274337,
      "grad_norm": 17.504682540893555,
      "learning_rate": 2.610619469026549e-05,
      "loss": 0.5666,
      "num_input_tokens_seen": 17904,
      "step": 60
    },
    {
      "epoch": 0.5752212389380531,
      "grad_norm": 7.692521095275879,
      "learning_rate": 2.831858407079646e-05,
      "loss": 0.2167,
      "num_input_tokens_seen": 19440,
      "step": 65
    },
    {
      "epoch": 0.6194690265486725,
      "grad_norm": 30.005857467651367,
      "learning_rate": 3.0530973451327434e-05,
      "loss": 0.4033,
      "num_input_tokens_seen": 21216,
      "step": 70
    },
    {
      "epoch": 0.6637168141592921,
      "grad_norm": 29.47793960571289,
      "learning_rate": 3.274336283185841e-05,
      "loss": 0.3693,
      "num_input_tokens_seen": 22400,
      "step": 75
    },
    {
      "epoch": 0.7079646017699115,
      "grad_norm": 9.944462776184082,
      "learning_rate": 3.495575221238938e-05,
      "loss": 0.2017,
      "num_input_tokens_seen": 24016,
      "step": 80
    },
    {
      "epoch": 0.7522123893805309,
      "grad_norm": 18.086078643798828,
      "learning_rate": 3.716814159292036e-05,
      "loss": 0.1915,
      "num_input_tokens_seen": 25360,
      "step": 85
    },
    {
      "epoch": 0.7964601769911505,
      "grad_norm": 97.63978576660156,
      "learning_rate": 3.938053097345133e-05,
      "loss": 1.1257,
      "num_input_tokens_seen": 26528,
      "step": 90
    },
    {
      "epoch": 0.8407079646017699,
      "grad_norm": 8.349373817443848,
      "learning_rate": 4.15929203539823e-05,
      "loss": 1.2739,
      "num_input_tokens_seen": 27728,
      "step": 95
    },
    {
      "epoch": 0.8849557522123894,
      "grad_norm": 3.84834885597229,
      "learning_rate": 4.380530973451328e-05,
      "loss": 0.2749,
      "num_input_tokens_seen": 29024,
      "step": 100
    },
    {
      "epoch": 0.9292035398230089,
      "grad_norm": 9.27602767944336,
      "learning_rate": 4.601769911504425e-05,
      "loss": 0.2006,
      "num_input_tokens_seen": 30128,
      "step": 105
    },
    {
      "epoch": 0.9734513274336283,
      "grad_norm": 6.60192346572876,
      "learning_rate": 4.823008849557522e-05,
      "loss": 0.3965,
      "num_input_tokens_seen": 31312,
      "step": 110
    },
    {
      "epoch": 1.008849557522124,
      "eval_loss": 0.9617744088172913,
      "eval_runtime": 0.7539,
      "eval_samples_per_second": 33.162,
      "eval_steps_per_second": 17.244,
      "num_input_tokens_seen": 32376,
      "step": 114
    },
    {
      "epoch": 1.0176991150442478,
      "grad_norm": 1.6416209936141968,
      "learning_rate": 4.9999880720033856e-05,
      "loss": 0.2448,
      "num_input_tokens_seen": 32664,
      "step": 115
    },
    {
      "epoch": 1.0619469026548674,
      "grad_norm": 6.446353435516357,
      "learning_rate": 4.999570604073014e-05,
      "loss": 0.6474,
      "num_input_tokens_seen": 33992,
      "step": 120
    },
    {
      "epoch": 1.1061946902654867,
      "grad_norm": 1.6060963869094849,
      "learning_rate": 4.998556850128691e-05,
      "loss": 0.217,
      "num_input_tokens_seen": 35528,
      "step": 125
    },
    {
      "epoch": 1.1504424778761062,
      "grad_norm": 18.64118003845215,
      "learning_rate": 4.996947052006874e-05,
      "loss": 0.6679,
      "num_input_tokens_seen": 36984,
      "step": 130
    },
    {
      "epoch": 1.1946902654867257,
      "grad_norm": 4.577888488769531,
      "learning_rate": 4.9947415937335635e-05,
      "loss": 0.4173,
      "num_input_tokens_seen": 38152,
      "step": 135
    },
    {
      "epoch": 1.238938053097345,
      "grad_norm": 4.134572982788086,
      "learning_rate": 4.9919410014326955e-05,
      "loss": 0.3019,
      "num_input_tokens_seen": 39528,
      "step": 140
    },
    {
      "epoch": 1.2831858407079646,
      "grad_norm": 10.661246299743652,
      "learning_rate": 4.98854594320063e-05,
      "loss": 0.3758,
      "num_input_tokens_seen": 41288,
      "step": 145
    },
    {
      "epoch": 1.3274336283185841,
      "grad_norm": 9.029864311218262,
      "learning_rate": 4.98455722894677e-05,
      "loss": 0.2423,
      "num_input_tokens_seen": 42744,
      "step": 150
    },
    {
      "epoch": 1.3716814159292037,
      "grad_norm": 9.982380867004395,
      "learning_rate": 4.979975810200359e-05,
      "loss": 0.7258,
      "num_input_tokens_seen": 43976,
      "step": 155
    },
    {
      "epoch": 1.415929203539823,
      "grad_norm": 2.6778085231781006,
      "learning_rate": 4.974802779883484e-05,
      "loss": 0.3808,
      "num_input_tokens_seen": 45576,
      "step": 160
    },
    {
      "epoch": 1.4601769911504425,
      "grad_norm": 1.8236883878707886,
      "learning_rate": 4.969039372050356e-05,
      "loss": 0.196,
      "num_input_tokens_seen": 47208,
      "step": 165
    },
    {
      "epoch": 1.504424778761062,
      "grad_norm": 0.23873771727085114,
      "learning_rate": 4.9626869615929175e-05,
      "loss": 0.3048,
      "num_input_tokens_seen": 48488,
      "step": 170
    },
    {
      "epoch": 1.5132743362831858,
      "eval_loss": 0.7283530831336975,
      "eval_runtime": 0.7557,
      "eval_samples_per_second": 33.084,
      "eval_steps_per_second": 17.204,
      "num_input_tokens_seen": 48728,
      "step": 171
    },
    {
      "epoch": 1.5486725663716814,
      "grad_norm": 3.6497254371643066,
      "learning_rate": 4.9557470639128554e-05,
      "loss": 0.9346,
      "num_input_tokens_seen": 49736,
      "step": 175
    },
    {
      "epoch": 1.592920353982301,
      "grad_norm": 1.4344072341918945,
      "learning_rate": 4.9482213345600936e-05,
      "loss": 0.2772,
      "num_input_tokens_seen": 50920,
      "step": 180
    },
    {
      "epoch": 1.6371681415929205,
      "grad_norm": 0.4830048680305481,
      "learning_rate": 4.940111568837854e-05,
      "loss": 0.3428,
      "num_input_tokens_seen": 52296,
      "step": 185
    },
    {
      "epoch": 1.6814159292035398,
      "grad_norm": 0.4717281758785248,
      "learning_rate": 4.931419701374377e-05,
      "loss": 0.1616,
      "num_input_tokens_seen": 54120,
      "step": 190
    },
    {
      "epoch": 1.7256637168141593,
      "grad_norm": 0.3464534878730774,
      "learning_rate": 4.922147805661402e-05,
      "loss": 0.1269,
      "num_input_tokens_seen": 55640,
      "step": 195
    },
    {
      "epoch": 1.7699115044247788,
      "grad_norm": 43.1662483215332,
      "learning_rate": 4.91229809355953e-05,
      "loss": 0.2072,
      "num_input_tokens_seen": 56728,
      "step": 200
    },
    {
      "epoch": 1.8141592920353982,
      "grad_norm": 0.1126592755317688,
      "learning_rate": 4.901872914770569e-05,
      "loss": 0.0061,
      "num_input_tokens_seen": 58344,
      "step": 205
    },
    {
      "epoch": 1.8584070796460177,
      "grad_norm": 0.09264620393514633,
      "learning_rate": 4.8908747562769995e-05,
      "loss": 0.4043,
      "num_input_tokens_seen": 59512,
      "step": 210
    },
    {
      "epoch": 1.9026548672566372,
      "grad_norm": 0.2140328735113144,
      "learning_rate": 4.8793062417486976e-05,
      "loss": 0.588,
      "num_input_tokens_seen": 60968,
      "step": 215
    },
    {
      "epoch": 1.9469026548672566,
      "grad_norm": 63.24400329589844,
      "learning_rate": 4.867170130917034e-05,
      "loss": 0.3411,
      "num_input_tokens_seen": 62152,
      "step": 220
    },
    {
      "epoch": 1.991150442477876,
      "grad_norm": 0.9429340362548828,
      "learning_rate": 4.8544693189165324e-05,
      "loss": 0.2021,
      "num_input_tokens_seen": 63432,
      "step": 225
    },
    {
      "epoch": 2.017699115044248,
      "eval_loss": 0.29189082980155945,
      "eval_runtime": 0.7555,
      "eval_samples_per_second": 33.092,
      "eval_steps_per_second": 17.208,
      "num_input_tokens_seen": 64040,
      "step": 228
    },
    {
      "epoch": 2.0353982300884956,
      "grad_norm": 0.1400822401046753,
      "learning_rate": 4.841206835594222e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 64520,
      "step": 230
    },
    {
      "epoch": 2.079646017699115,
      "grad_norm": 0.05922820046544075,
      "learning_rate": 4.8273858447868424e-05,
      "loss": 0.0191,
      "num_input_tokens_seen": 66104,
      "step": 235
    },
    {
      "epoch": 2.1238938053097347,
      "grad_norm": 0.10322703421115875,
      "learning_rate": 4.813009643566101e-05,
      "loss": 0.003,
      "num_input_tokens_seen": 67464,
      "step": 240
    },
    {
      "epoch": 2.168141592920354,
      "grad_norm": 0.48469221591949463,
      "learning_rate": 4.798081661452135e-05,
      "loss": 0.188,
      "num_input_tokens_seen": 68824,
      "step": 245
    },
    {
      "epoch": 2.2123893805309733,
      "grad_norm": 0.1246049553155899,
      "learning_rate": 4.7826054595953815e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 70040,
      "step": 250
    },
    {
      "epoch": 2.256637168141593,
      "grad_norm": 11.989933013916016,
      "learning_rate": 4.766584729927049e-05,
      "loss": 0.5932,
      "num_input_tokens_seen": 71784,
      "step": 255
    },
    {
      "epoch": 2.3008849557522124,
      "grad_norm": 18.2181339263916,
      "learning_rate": 4.750023294278381e-05,
      "loss": 0.6066,
      "num_input_tokens_seen": 73096,
      "step": 260
    },
    {
      "epoch": 2.3451327433628317,
      "grad_norm": 9.733284950256348,
      "learning_rate": 4.732925103468944e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 74328,
      "step": 265
    },
    {
      "epoch": 2.3893805309734515,
      "grad_norm": 1.9108598232269287,
      "learning_rate": 4.715294236364135e-05,
      "loss": 0.2008,
      "num_input_tokens_seen": 75656,
      "step": 270
    },
    {
      "epoch": 2.433628318584071,
      "grad_norm": 7.191739559173584,
      "learning_rate": 4.6971348989021446e-05,
      "loss": 0.2102,
      "num_input_tokens_seen": 77016,
      "step": 275
    },
    {
      "epoch": 2.47787610619469,
      "grad_norm": 11.626720428466797,
      "learning_rate": 4.6784514230906146e-05,
      "loss": 0.4873,
      "num_input_tokens_seen": 78376,
      "step": 280
    },
    {
      "epoch": 2.52212389380531,
      "grad_norm": 34.25971221923828,
      "learning_rate": 4.659248265973205e-05,
      "loss": 0.4291,
      "num_input_tokens_seen": 79784,
      "step": 285
    },
    {
      "epoch": 2.52212389380531,
      "eval_loss": 0.8385664224624634,
      "eval_runtime": 0.9317,
      "eval_samples_per_second": 26.833,
      "eval_steps_per_second": 13.953,
      "num_input_tokens_seen": 79784,
      "step": 285
    },
    {
      "epoch": 2.566371681415929,
      "grad_norm": 8.708770751953125,
      "learning_rate": 4.639530008566349e-05,
      "loss": 0.8388,
      "num_input_tokens_seen": 81192,
      "step": 290
    },
    {
      "epoch": 2.6106194690265485,
      "grad_norm": 0.19684574007987976,
      "learning_rate": 4.6193013547664255e-05,
      "loss": 0.0309,
      "num_input_tokens_seen": 82552,
      "step": 295
    },
    {
      "epoch": 2.6548672566371683,
      "grad_norm": 3.585326910018921,
      "learning_rate": 4.5985671302276165e-05,
      "loss": 0.6214,
      "num_input_tokens_seen": 84280,
      "step": 300
    },
    {
      "epoch": 2.6991150442477876,
      "grad_norm": 0.8783462643623352,
      "learning_rate": 4.577332281210727e-05,
      "loss": 0.1596,
      "num_input_tokens_seen": 85752,
      "step": 305
    },
    {
      "epoch": 2.7433628318584073,
      "grad_norm": 0.13845697045326233,
      "learning_rate": 4.55560187340322e-05,
      "loss": 0.3166,
      "num_input_tokens_seen": 87448,
      "step": 310
    },
    {
      "epoch": 2.7876106194690267,
      "grad_norm": 2.781928777694702,
      "learning_rate": 4.533381090710776e-05,
      "loss": 0.4923,
      "num_input_tokens_seen": 89080,
      "step": 315
    },
    {
      "epoch": 2.831858407079646,
      "grad_norm": 7.809700965881348,
      "learning_rate": 4.5106752340206435e-05,
      "loss": 0.2046,
      "num_input_tokens_seen": 90456,
      "step": 320
    },
    {
      "epoch": 2.8761061946902657,
      "grad_norm": 0.18478740751743317,
      "learning_rate": 4.4874897199370805e-05,
      "loss": 0.7264,
      "num_input_tokens_seen": 91624,
      "step": 325
    },
    {
      "epoch": 2.920353982300885,
      "grad_norm": 0.2102925181388855,
      "learning_rate": 4.463830079489196e-05,
      "loss": 0.1935,
      "num_input_tokens_seen": 92792,
      "step": 330
    },
    {
      "epoch": 2.9646017699115044,
      "grad_norm": 10.054224967956543,
      "learning_rate": 4.4397019568114987e-05,
      "loss": 0.2797,
      "num_input_tokens_seen": 94280,
      "step": 335
    },
    {
      "epoch": 3.0088495575221237,
      "grad_norm": 0.1494530737400055,
      "learning_rate": 4.415111107797445e-05,
      "loss": 0.0067,
      "num_input_tokens_seen": 95752,
      "step": 340
    },
    {
      "epoch": 3.0265486725663715,
      "eval_loss": 0.3567735254764557,
      "eval_runtime": 0.751,
      "eval_samples_per_second": 33.29,
      "eval_steps_per_second": 17.311,
      "num_input_tokens_seen": 96200,
      "step": 342
    },
    {
      "epoch": 3.0530973451327434,
      "grad_norm": 0.04667724668979645,
      "learning_rate": 4.390063398726356e-05,
      "loss": 0.314,
      "num_input_tokens_seen": 97000,
      "step": 345
    },
    {
      "epoch": 3.0973451327433628,
      "grad_norm": 0.11600804328918457,
      "learning_rate": 4.3645648048639734e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 98248,
      "step": 350
    },
    {
      "epoch": 3.1415929203539825,
      "grad_norm": 0.2308310866355896,
      "learning_rate": 4.338621409037031e-05,
      "loss": 0.193,
      "num_input_tokens_seen": 99800,
      "step": 355
    },
    {
      "epoch": 3.185840707964602,
      "grad_norm": 6.516209125518799,
      "learning_rate": 4.312239400182166e-05,
      "loss": 0.1346,
      "num_input_tokens_seen": 101080,
      "step": 360
    },
    {
      "epoch": 3.230088495575221,
      "grad_norm": 1.2837355136871338,
      "learning_rate": 4.285425071869511e-05,
      "loss": 0.173,
      "num_input_tokens_seen": 102488,
      "step": 365
    },
    {
      "epoch": 3.274336283185841,
      "grad_norm": 1.1854627132415771,
      "learning_rate": 4.258184820801331e-05,
      "loss": 0.0875,
      "num_input_tokens_seen": 103912,
      "step": 370
    },
    {
      "epoch": 3.3185840707964602,
      "grad_norm": 1.033186912536621,
      "learning_rate": 4.230525145286057e-05,
      "loss": 0.0617,
      "num_input_tokens_seen": 105400,
      "step": 375
    },
    {
      "epoch": 3.3628318584070795,
      "grad_norm": 0.08329396694898605,
      "learning_rate": 4.2024526436880785e-05,
      "loss": 0.0255,
      "num_input_tokens_seen": 106920,
      "step": 380
    },
    {
      "epoch": 3.4070796460176993,
      "grad_norm": 9.243614196777344,
      "learning_rate": 4.173974012853665e-05,
      "loss": 0.4644,
      "num_input_tokens_seen": 108280,
      "step": 385
    },
    {
      "epoch": 3.4513274336283186,
      "grad_norm": 7.288797855377197,
      "learning_rate": 4.1450960465134025e-05,
      "loss": 0.5902,
      "num_input_tokens_seen": 110024,
      "step": 390
    },
    {
      "epoch": 3.495575221238938,
      "grad_norm": 0.027472253888845444,
      "learning_rate": 4.1158256336615096e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 111304,
      "step": 395
    },
    {
      "epoch": 3.5309734513274336,
      "eval_loss": 0.32799670100212097,
      "eval_runtime": 0.7498,
      "eval_samples_per_second": 33.344,
      "eval_steps_per_second": 17.339,
      "num_input_tokens_seen": 112440,
      "step": 399
    },
    {
      "epoch": 3.5398230088495577,
      "grad_norm": 4.507176876068115,
      "learning_rate": 4.086169756912431e-05,
      "loss": 0.4656,
      "num_input_tokens_seen": 112696,
      "step": 400
    },
    {
      "epoch": 3.584070796460177,
      "grad_norm": 0.3329651355743408,
      "learning_rate": 4.056135490835098e-05,
      "loss": 0.2085,
      "num_input_tokens_seen": 114248,
      "step": 405
    },
    {
      "epoch": 3.6283185840707963,
      "grad_norm": 1.9532846212387085,
      "learning_rate": 4.025730000265251e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 115400,
      "step": 410
    },
    {
      "epoch": 3.672566371681416,
      "grad_norm": 2.9293394088745117,
      "learning_rate": 3.994960538596232e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 116808,
      "step": 415
    },
    {
      "epoch": 3.7168141592920354,
      "grad_norm": 0.6144419312477112,
      "learning_rate": 3.963834446048644e-05,
      "loss": 0.4318,
      "num_input_tokens_seen": 118056,
      "step": 420
    },
    {
      "epoch": 3.7610619469026547,
      "grad_norm": 0.2290184199810028,
      "learning_rate": 3.9323591479193074e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 119512,
      "step": 425
    },
    {
      "epoch": 3.8053097345132745,
      "grad_norm": 8.022658348083496,
      "learning_rate": 3.90054215280991e-05,
      "loss": 0.2427,
      "num_input_tokens_seen": 120872,
      "step": 430
    },
    {
      "epoch": 3.849557522123894,
      "grad_norm": 2.313319206237793,
      "learning_rate": 3.8683910508357934e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 122376,
      "step": 435
    },
    {
      "epoch": 3.893805309734513,
      "grad_norm": 0.4085032343864441,
      "learning_rate": 3.83591351181529e-05,
      "loss": 0.1838,
      "num_input_tokens_seen": 123544,
      "step": 440
    },
    {
      "epoch": 3.938053097345133,
      "grad_norm": 1.2104312181472778,
      "learning_rate": 3.803117283440044e-05,
      "loss": 0.196,
      "num_input_tokens_seen": 125720,
      "step": 445
    },
    {
      "epoch": 3.982300884955752,
      "grad_norm": 0.8213343024253845,
      "learning_rate": 3.7700101894267615e-05,
      "loss": 0.3714,
      "num_input_tokens_seen": 126952,
      "step": 450
    },
    {
      "epoch": 4.0265486725663715,
      "grad_norm": 0.8457050919532776,
      "learning_rate": 3.736600127650817e-05,
      "loss": 0.1948,
      "num_input_tokens_seen": 128488,
      "step": 455
    },
    {
      "epoch": 4.035398230088496,
      "eval_loss": 0.24273568391799927,
      "eval_runtime": 0.7503,
      "eval_samples_per_second": 33.32,
      "eval_steps_per_second": 17.327,
      "num_input_tokens_seen": 128712,
      "step": 456
    },
    {
      "epoch": 4.070796460176991,
      "grad_norm": 0.8174092769622803,
      "learning_rate": 3.702895068262177e-05,
      "loss": 0.06,
      "num_input_tokens_seen": 129880,
      "step": 460
    },
    {
      "epoch": 4.115044247787611,
      "grad_norm": 4.760356426239014,
      "learning_rate": 3.66890305178407e-05,
      "loss": 0.1748,
      "num_input_tokens_seen": 131176,
      "step": 465
    },
    {
      "epoch": 4.15929203539823,
      "grad_norm": 0.1795598864555359,
      "learning_rate": 3.63463218719489e-05,
      "loss": 0.1834,
      "num_input_tokens_seen": 132376,
      "step": 470
    },
    {
      "epoch": 4.20353982300885,
      "grad_norm": 0.09256363660097122,
      "learning_rate": 3.600090649993741e-05,
      "loss": 0.2312,
      "num_input_tokens_seen": 133608,
      "step": 475
    },
    {
      "epoch": 4.247787610619469,
      "grad_norm": 0.24030235409736633,
      "learning_rate": 3.565286680250138e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 135080,
      "step": 480
    },
    {
      "epoch": 4.292035398230088,
      "grad_norm": 0.6925850510597229,
      "learning_rate": 3.530228580638291e-05,
      "loss": 0.0585,
      "num_input_tokens_seen": 136264,
      "step": 485
    },
    {
      "epoch": 4.336283185840708,
      "grad_norm": 0.26487845182418823,
      "learning_rate": 3.494924714456454e-05,
      "loss": 0.2241,
      "num_input_tokens_seen": 137480,
      "step": 490
    },
    {
      "epoch": 4.380530973451328,
      "grad_norm": 5.6674017906188965,
      "learning_rate": 3.459383503631823e-05,
      "loss": 0.3258,
      "num_input_tokens_seen": 138536,
      "step": 495
    },
    {
      "epoch": 4.424778761061947,
      "grad_norm": 0.5745865702629089,
      "learning_rate": 3.423613426711431e-05,
      "loss": 0.0396,
      "num_input_tokens_seen": 140360,
      "step": 500
    },
    {
      "epoch": 4.469026548672566,
      "grad_norm": 0.23020365834236145,
      "learning_rate": 3.3876230168395525e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 141784,
      "step": 505
    },
    {
      "epoch": 4.513274336283186,
      "grad_norm": 2.4007861614227295,
      "learning_rate": 3.3514208597220705e-05,
      "loss": 0.1894,
      "num_input_tokens_seen": 143192,
      "step": 510
    },
    {
      "epoch": 4.539823008849558,
      "eval_loss": 0.47883257269859314,
      "eval_runtime": 0.7472,
      "eval_samples_per_second": 33.46,
      "eval_steps_per_second": 17.399,
      "num_input_tokens_seen": 143944,
      "step": 513
    },
    {
      "epoch": 4.557522123893805,
      "grad_norm": 9.113517761230469,
      "learning_rate": 3.315015591578314e-05,
      "loss": 0.1333,
      "num_input_tokens_seen": 144440,
      "step": 515
    },
    {
      "epoch": 4.601769911504425,
      "grad_norm": 0.03672410547733307,
      "learning_rate": 3.278415897080839e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 145768,
      "step": 520
    },
    {
      "epoch": 4.646017699115045,
      "grad_norm": 2.4386110305786133,
      "learning_rate": 3.2416305072836557e-05,
      "loss": 0.2245,
      "num_input_tokens_seen": 147128,
      "step": 525
    },
    {
      "epoch": 4.6902654867256635,
      "grad_norm": 0.5586014986038208,
      "learning_rate": 3.204668197539385e-05,
      "loss": 0.0162,
      "num_input_tokens_seen": 148184,
      "step": 530
    },
    {
      "epoch": 4.734513274336283,
      "grad_norm": 11.894312858581543,
      "learning_rate": 3.167537785405854e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 149880,
      "step": 535
    },
    {
      "epoch": 4.778761061946903,
      "grad_norm": 1.8908299207687378,
      "learning_rate": 3.13024812854262e-05,
      "loss": 0.3182,
      "num_input_tokens_seen": 151528,
      "step": 540
    },
    {
      "epoch": 4.823008849557522,
      "grad_norm": 0.36227720975875854,
      "learning_rate": 3.092808122597924e-05,
      "loss": 0.2819,
      "num_input_tokens_seen": 153080,
      "step": 545
    },
    {
      "epoch": 4.867256637168142,
      "grad_norm": 0.4420020878314972,
      "learning_rate": 3.0552266990865934e-05,
      "loss": 0.0733,
      "num_input_tokens_seen": 154952,
      "step": 550
    },
    {
      "epoch": 4.911504424778761,
      "grad_norm": 0.13723275065422058,
      "learning_rate": 3.017512823259373e-05,
      "loss": 0.1868,
      "num_input_tokens_seen": 156136,
      "step": 555
    },
    {
      "epoch": 4.95575221238938,
      "grad_norm": 0.38662219047546387,
      "learning_rate": 2.9796754919642228e-05,
      "loss": 0.0063,
      "num_input_tokens_seen": 157560,
      "step": 560
    },
    {
      "epoch": 5.0,
      "grad_norm": 8.381264686584473,
      "learning_rate": 2.941723731500068e-05,
      "loss": 0.1084,
      "num_input_tokens_seen": 158768,
      "step": 565
    },
    {
      "epoch": 5.04424778761062,
      "grad_norm": 0.5755460262298584,
      "learning_rate": 2.9036665954635266e-05,
      "loss": 0.2664,
      "num_input_tokens_seen": 160016,
      "step": 570
    },
    {
      "epoch": 5.04424778761062,
      "eval_loss": 0.3219310939311981,
      "eval_runtime": 0.7495,
      "eval_samples_per_second": 33.356,
      "eval_steps_per_second": 17.345,
      "num_input_tokens_seen": 160016,
      "step": 570
    },
    {
      "epoch": 5.088495575221239,
      "grad_norm": 0.21571117639541626,
      "learning_rate": 2.8655131625891274e-05,
      "loss": 0.1232,
      "num_input_tokens_seen": 161360,
      "step": 575
    },
    {
      "epoch": 5.132743362831858,
      "grad_norm": 0.5637882351875305,
      "learning_rate": 2.8272725345835278e-05,
      "loss": 0.0079,
      "num_input_tokens_seen": 163104,
      "step": 580
    },
    {
      "epoch": 5.176991150442478,
      "grad_norm": 2.5816102027893066,
      "learning_rate": 2.7889538339542527e-05,
      "loss": 0.1038,
      "num_input_tokens_seen": 164368,
      "step": 585
    },
    {
      "epoch": 5.221238938053097,
      "grad_norm": 0.09894769638776779,
      "learning_rate": 2.7505662018334698e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 165936,
      "step": 590
    },
    {
      "epoch": 5.265486725663717,
      "grad_norm": 4.558467388153076,
      "learning_rate": 2.712118795797329e-05,
      "loss": 0.2798,
      "num_input_tokens_seen": 167360,
      "step": 595
    },
    {
      "epoch": 5.3097345132743365,
      "grad_norm": 0.6896599531173706,
      "learning_rate": 2.6736207876813646e-05,
      "loss": 0.1178,
      "num_input_tokens_seen": 168944,
      "step": 600
    },
    {
      "epoch": 5.353982300884955,
      "grad_norm": 0.8930271863937378,
      "learning_rate": 2.635081361392513e-05,
      "loss": 0.1037,
      "num_input_tokens_seen": 170672,
      "step": 605
    },
    {
      "epoch": 5.398230088495575,
      "grad_norm": 0.026197021827101707,
      "learning_rate": 2.596509710718235e-05,
      "loss": 0.0214,
      "num_input_tokens_seen": 172016,
      "step": 610
    },
    {
      "epoch": 5.442477876106195,
      "grad_norm": 0.28106260299682617,
      "learning_rate": 2.5579150371332956e-05,
      "loss": 0.0438,
      "num_input_tokens_seen": 173184,
      "step": 615
    },
    {
      "epoch": 5.486725663716814,
      "grad_norm": 0.03865925967693329,
      "learning_rate": 2.5193065476046955e-05,
      "loss": 0.0007,
      "num_input_tokens_seen": 174352,
      "step": 620
    },
    {
      "epoch": 5.530973451327434,
      "grad_norm": 0.31933000683784485,
      "learning_rate": 2.480693452395305e-05,
      "loss": 0.0356,
      "num_input_tokens_seen": 175968,
      "step": 625
    },
    {
      "epoch": 5.548672566371682,
      "eval_loss": 0.44988709688186646,
      "eval_runtime": 0.7511,
      "eval_samples_per_second": 33.285,
      "eval_steps_per_second": 17.308,
      "num_input_tokens_seen": 176688,
      "step": 627
    },
    {
      "epoch": 5.575221238938053,
      "grad_norm": 0.42749717831611633,
      "learning_rate": 2.4420849628667046e-05,
      "loss": 0.1774,
      "num_input_tokens_seen": 177408,
      "step": 630
    },
    {
      "epoch": 5.619469026548672,
      "grad_norm": 0.013716431334614754,
      "learning_rate": 2.4034902892817648e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 178608,
      "step": 635
    },
    {
      "epoch": 5.663716814159292,
      "grad_norm": 0.009475010447204113,
      "learning_rate": 2.3649186386074872e-05,
      "loss": 0.0385,
      "num_input_tokens_seen": 179824,
      "step": 640
    },
    {
      "epoch": 5.707964601769912,
      "grad_norm": 0.3431243300437927,
      "learning_rate": 2.3263792123186353e-05,
      "loss": 0.2638,
      "num_input_tokens_seen": 181088,
      "step": 645
    },
    {
      "epoch": 5.752212389380531,
      "grad_norm": 8.075549125671387,
      "learning_rate": 2.2878812042026713e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 182816,
      "step": 650
    },
    {
      "epoch": 5.79646017699115,
      "grad_norm": 6.323182582855225,
      "learning_rate": 2.2494337981665304e-05,
      "loss": 0.1909,
      "num_input_tokens_seen": 184224,
      "step": 655
    },
    {
      "epoch": 5.84070796460177,
      "grad_norm": 1.3228939771652222,
      "learning_rate": 2.2110461660457482e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 185776,
      "step": 660
    },
    {
      "epoch": 5.88495575221239,
      "grad_norm": 0.452834814786911,
      "learning_rate": 2.1727274654164728e-05,
      "loss": 0.1154,
      "num_input_tokens_seen": 187184,
      "step": 665
    },
    {
      "epoch": 5.929203539823009,
      "grad_norm": 0.7257612347602844,
      "learning_rate": 2.1344868374108732e-05,
      "loss": 0.2587,
      "num_input_tokens_seen": 188528,
      "step": 670
    },
    {
      "epoch": 5.9734513274336285,
      "grad_norm": 7.712823867797852,
      "learning_rate": 2.096333404536474e-05,
      "loss": 0.07,
      "num_input_tokens_seen": 189680,
      "step": 675
    },
    {
      "epoch": 6.017699115044247,
      "grad_norm": 0.3695467412471771,
      "learning_rate": 2.0582762684999327e-05,
      "loss": 0.0116,
      "num_input_tokens_seen": 191008,
      "step": 680
    },
    {
      "epoch": 6.053097345132743,
      "eval_loss": 0.28231462836265564,
      "eval_runtime": 0.9304,
      "eval_samples_per_second": 26.87,
      "eval_steps_per_second": 13.972,
      "num_input_tokens_seen": 192272,
      "step": 684
    },
    {
      "epoch": 6.061946902654867,
      "grad_norm": 11.890759468078613,
      "learning_rate": 2.0203245080357775e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 192512,
      "step": 685
    },
    {
      "epoch": 6.106194690265487,
      "grad_norm": 4.551051139831543,
      "learning_rate": 1.982487176740627e-05,
      "loss": 0.0633,
      "num_input_tokens_seen": 193904,
      "step": 690
    },
    {
      "epoch": 6.150442477876107,
      "grad_norm": 0.01000402681529522,
      "learning_rate": 1.944773300913407e-05,
      "loss": 0.3281,
      "num_input_tokens_seen": 195616,
      "step": 695
    },
    {
      "epoch": 6.1946902654867255,
      "grad_norm": 0.009216882288455963,
      "learning_rate": 1.907191877402076e-05,
      "loss": 0.0028,
      "num_input_tokens_seen": 196832,
      "step": 700
    },
    {
      "epoch": 6.238938053097345,
      "grad_norm": 0.11673197150230408,
      "learning_rate": 1.8697518714573807e-05,
      "loss": 0.0705,
      "num_input_tokens_seen": 198592,
      "step": 705
    },
    {
      "epoch": 6.283185840707965,
      "grad_norm": 0.30128544569015503,
      "learning_rate": 1.832462214594146e-05,
      "loss": 0.6563,
      "num_input_tokens_seen": 199664,
      "step": 710
    },
    {
      "epoch": 6.327433628318584,
      "grad_norm": 0.12938998639583588,
      "learning_rate": 1.7953318024606157e-05,
      "loss": 0.0088,
      "num_input_tokens_seen": 201456,
      "step": 715
    },
    {
      "epoch": 6.371681415929204,
      "grad_norm": 5.580306053161621,
      "learning_rate": 1.7583694927163453e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 202752,
      "step": 720
    },
    {
      "epoch": 6.415929203539823,
      "grad_norm": 0.06375036388635635,
      "learning_rate": 1.7215841029191616e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 204336,
      "step": 725
    },
    {
      "epoch": 6.460176991150442,
      "grad_norm": 6.705235481262207,
      "learning_rate": 1.6849844084216874e-05,
      "loss": 0.013,
      "num_input_tokens_seen": 205488,
      "step": 730
    },
    {
      "epoch": 6.504424778761062,
      "grad_norm": 0.0064069428481161594,
      "learning_rate": 1.648579140277931e-05,
      "loss": 0.0048,
      "num_input_tokens_seen": 207008,
      "step": 735
    },
    {
      "epoch": 6.548672566371682,
      "grad_norm": 0.006996363867074251,
      "learning_rate": 1.612376983160449e-05,
      "loss": 0.0005,
      "num_input_tokens_seen": 208512,
      "step": 740
    },
    {
      "epoch": 6.557522123893805,
      "eval_loss": 0.4587627649307251,
      "eval_runtime": 0.7495,
      "eval_samples_per_second": 33.355,
      "eval_steps_per_second": 17.345,
      "num_input_tokens_seen": 208944,
      "step": 741
    },
    {
      "epoch": 6.592920353982301,
      "grad_norm": 0.05237611010670662,
      "learning_rate": 1.576386573288569e-05,
      "loss": 0.0873,
      "num_input_tokens_seen": 209888,
      "step": 745
    },
    {
      "epoch": 6.6371681415929205,
      "grad_norm": 0.011837458238005638,
      "learning_rate": 1.5406164963681774e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 211008,
      "step": 750
    },
    {
      "epoch": 6.68141592920354,
      "grad_norm": 0.006387337576597929,
      "learning_rate": 1.5050752855435457e-05,
      "loss": 0.0682,
      "num_input_tokens_seen": 212240,
      "step": 755
    },
    {
      "epoch": 6.725663716814159,
      "grad_norm": 37.566978454589844,
      "learning_rate": 1.4697714193617097e-05,
      "loss": 0.1893,
      "num_input_tokens_seen": 213456,
      "step": 760
    },
    {
      "epoch": 6.769911504424779,
      "grad_norm": 0.006813609041273594,
      "learning_rate": 1.4347133197498619e-05,
      "loss": 0.0019,
      "num_input_tokens_seen": 214768,
      "step": 765
    },
    {
      "epoch": 6.814159292035399,
      "grad_norm": 0.13870197534561157,
      "learning_rate": 1.3999093500062587e-05,
      "loss": 0.0011,
      "num_input_tokens_seen": 216000,
      "step": 770
    },
    {
      "epoch": 6.8584070796460175,
      "grad_norm": 0.0034726073499768972,
      "learning_rate": 1.3653678128051106e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 217520,
      "step": 775
    },
    {
      "epoch": 6.902654867256637,
      "grad_norm": 0.2616535723209381,
      "learning_rate": 1.3310969482159297e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 219040,
      "step": 780
    },
    {
      "epoch": 6.946902654867257,
      "grad_norm": 0.018871121108531952,
      "learning_rate": 1.2971049317378237e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 220688,
      "step": 785
    },
    {
      "epoch": 6.991150442477876,
      "grad_norm": 0.005359095521271229,
      "learning_rate": 1.263399872349183e-05,
      "loss": 0.0008,
      "num_input_tokens_seen": 222304,
      "step": 790
    },
    {
      "epoch": 7.035398230088496,
      "grad_norm": 0.05084985867142677,
      "learning_rate": 1.2299898105732385e-05,
      "loss": 0.0509,
      "num_input_tokens_seen": 223584,
      "step": 795
    },
    {
      "epoch": 7.061946902654867,
      "eval_loss": 0.34447139501571655,
      "eval_runtime": 0.9229,
      "eval_samples_per_second": 27.088,
      "eval_steps_per_second": 14.086,
      "num_input_tokens_seen": 224288,
      "step": 798
    },
    {
      "epoch": 7.079646017699115,
      "grad_norm": 0.006537236738950014,
      "learning_rate": 1.1968827165599564e-05,
      "loss": 0.0036,
      "num_input_tokens_seen": 224816,
      "step": 800
    },
    {
      "epoch": 7.123893805309734,
      "grad_norm": 0.0038681919686496258,
      "learning_rate": 1.1640864881847105e-05,
      "loss": 0.0405,
      "num_input_tokens_seen": 226192,
      "step": 805
    },
    {
      "epoch": 7.168141592920354,
      "grad_norm": 0.00919914711266756,
      "learning_rate": 1.1316089491642076e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 227440,
      "step": 810
    },
    {
      "epoch": 7.212389380530974,
      "grad_norm": 0.0037935979198664427,
      "learning_rate": 1.099457847190091e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 228928,
      "step": 815
    },
    {
      "epoch": 7.256637168141593,
      "grad_norm": 0.004989915061742067,
      "learning_rate": 1.067640852080693e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 230416,
      "step": 820
    },
    {
      "epoch": 7.300884955752212,
      "grad_norm": 0.004112043417990208,
      "learning_rate": 1.0361655539513565e-05,
      "loss": 0.0665,
      "num_input_tokens_seen": 231536,
      "step": 825
    },
    {
      "epoch": 7.345132743362832,
      "grad_norm": 0.06868399679660797,
      "learning_rate": 1.0050394614037687e-05,
      "loss": 0.2363,
      "num_input_tokens_seen": 233072,
      "step": 830
    },
    {
      "epoch": 7.389380530973451,
      "grad_norm": 0.022683456540107727,
      "learning_rate": 9.742699997347498e-06,
      "loss": 0.0006,
      "num_input_tokens_seen": 234416,
      "step": 835
    },
    {
      "epoch": 7.433628318584071,
      "grad_norm": 0.17499703168869019,
      "learning_rate": 9.438645091649029e-06,
      "loss": 0.0015,
      "num_input_tokens_seen": 235776,
      "step": 840
    },
    {
      "epoch": 7.477876106194691,
      "grad_norm": 0.10602128505706787,
      "learning_rate": 9.138302430875695e-06,
      "loss": 0.0007,
      "num_input_tokens_seen": 237248,
      "step": 845
    },
    {
      "epoch": 7.522123893805309,
      "grad_norm": 0.2794215679168701,
      "learning_rate": 8.841743663384913e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 238544,
      "step": 850
    },
    {
      "epoch": 7.566371681415929,
      "grad_norm": 0.03861619532108307,
      "learning_rate": 8.549039534865979e-06,
      "loss": 0.0031,
      "num_input_tokens_seen": 239840,
      "step": 855
    },
    {
      "epoch": 7.566371681415929,
      "eval_loss": 0.4802214503288269,
      "eval_runtime": 0.7493,
      "eval_samples_per_second": 33.364,
      "eval_steps_per_second": 17.349,
      "num_input_tokens_seen": 239840,
      "step": 855
    },
    {
      "epoch": 7.610619469026549,
      "grad_norm": 0.010991969145834446,
      "learning_rate": 8.260259871463352e-06,
      "loss": 0.0683,
      "num_input_tokens_seen": 241440,
      "step": 860
    },
    {
      "epoch": 7.654867256637168,
      "grad_norm": 0.09714938700199127,
      "learning_rate": 7.975473563119216e-06,
      "loss": 0.0013,
      "num_input_tokens_seen": 243216,
      "step": 865
    },
    {
      "epoch": 7.699115044247788,
      "grad_norm": 3.762711524963379,
      "learning_rate": 7.69474854713943e-06,
      "loss": 0.0118,
      "num_input_tokens_seen": 244400,
      "step": 870
    },
    {
      "epoch": 7.743362831858407,
      "grad_norm": 0.029518570750951767,
      "learning_rate": 7.4181517919866914e-06,
      "loss": 0.0035,
      "num_input_tokens_seen": 245520,
      "step": 875
    },
    {
      "epoch": 7.787610619469026,
      "grad_norm": 0.07981577515602112,
      "learning_rate": 7.1457492813049e-06,
      "loss": 0.0007,
      "num_input_tokens_seen": 246688,
      "step": 880
    },
    {
      "epoch": 7.831858407079646,
      "grad_norm": 0.1901676207780838,
      "learning_rate": 6.877605998178344e-06,
      "loss": 0.0036,
      "num_input_tokens_seen": 248320,
      "step": 885
    },
    {
      "epoch": 7.876106194690266,
      "grad_norm": 0.053983792662620544,
      "learning_rate": 6.613785909629686e-06,
      "loss": 0.0082,
      "num_input_tokens_seen": 249616,
      "step": 890
    },
    {
      "epoch": 7.920353982300885,
      "grad_norm": 0.04075269773602486,
      "learning_rate": 6.354351951360268e-06,
      "loss": 0.0004,
      "num_input_tokens_seen": 251264,
      "step": 895
    },
    {
      "epoch": 7.964601769911504,
      "grad_norm": 0.007756324484944344,
      "learning_rate": 6.099366012736438e-06,
      "loss": 0.0025,
      "num_input_tokens_seen": 252768,
      "step": 900
    },
    {
      "epoch": 8.008849557522124,
      "grad_norm": 0.017194686457514763,
      "learning_rate": 5.848888922025553e-06,
      "loss": 0.0002,
      "num_input_tokens_seen": 254096,
      "step": 905
    },
    {
      "epoch": 8.053097345132743,
      "grad_norm": 0.005764801520854235,
      "learning_rate": 5.602980431885024e-06,
      "loss": 0.0002,
      "num_input_tokens_seen": 255408,
      "step": 910
    },
    {
      "epoch": 8.070796460176991,
      "eval_loss": 0.4264009892940521,
      "eval_runtime": 0.7482,
      "eval_samples_per_second": 33.411,
      "eval_steps_per_second": 17.374,
      "num_input_tokens_seen": 255984,
      "step": 912
    },
    {
      "epoch": 8.097345132743364,
      "grad_norm": 0.003981693647801876,
      "learning_rate": 5.361699205108042e-06,
      "loss": 0.0892,
      "num_input_tokens_seen": 257008,
      "step": 915
    },
    {
      "epoch": 8.141592920353983,
      "grad_norm": 0.032894838601350784,
      "learning_rate": 5.125102800629203e-06,
      "loss": 0.0049,
      "num_input_tokens_seen": 258176,
      "step": 920
    },
    {
      "epoch": 8.185840707964601,
      "grad_norm": 0.007304678205400705,
      "learning_rate": 4.893247659793568e-06,
      "loss": 0.0002,
      "num_input_tokens_seen": 259552,
      "step": 925
    },
    {
      "epoch": 8.230088495575222,
      "grad_norm": 0.19905464351177216,
      "learning_rate": 4.666189092892245e-06,
      "loss": 0.0005,
      "num_input_tokens_seen": 261104,
      "step": 930
    },
    {
      "epoch": 8.274336283185841,
      "grad_norm": 0.020822672173380852,
      "learning_rate": 4.443981265967806e-06,
      "loss": 0.0224,
      "num_input_tokens_seen": 262496,
      "step": 935
    },
    {
      "epoch": 8.31858407079646,
      "grad_norm": 0.2635118365287781,
      "learning_rate": 4.226677187892739e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 263744,
      "step": 940
    },
    {
      "epoch": 8.36283185840708,
      "grad_norm": 0.11042467504739761,
      "learning_rate": 4.014328697723835e-06,
      "loss": 0.0004,
      "num_input_tokens_seen": 265024,
      "step": 945
    },
    {
      "epoch": 8.4070796460177,
      "grad_norm": 0.005107547156512737,
      "learning_rate": 3.806986452335748e-06,
      "loss": 0.09,
      "num_input_tokens_seen": 266320,
      "step": 950
    },
    {
      "epoch": 8.451327433628318,
      "grad_norm": 0.26767849922180176,
      "learning_rate": 3.604699914336515e-06,
      "loss": 0.0009,
      "num_input_tokens_seen": 267648,
      "step": 955
    },
    {
      "epoch": 8.495575221238939,
      "grad_norm": 0.035442568361759186,
      "learning_rate": 3.4075173402679574e-06,
      "loss": 0.0005,
      "num_input_tokens_seen": 269264,
      "step": 960
    },
    {
      "epoch": 8.539823008849558,
      "grad_norm": 0.006228589452803135,
      "learning_rate": 3.215485769093862e-06,
      "loss": 0.0014,
      "num_input_tokens_seen": 270816,
      "step": 965
    },
    {
      "epoch": 8.575221238938052,
      "eval_loss": 0.36992689967155457,
      "eval_runtime": 0.7484,
      "eval_samples_per_second": 33.403,
      "eval_steps_per_second": 17.37,
      "num_input_tokens_seen": 272064,
      "step": 969
    },
    {
      "epoch": 8.584070796460177,
      "grad_norm": 0.07714986801147461,
      "learning_rate": 3.028651010978556e-06,
      "loss": 0.0008,
      "num_input_tokens_seen": 272304,
      "step": 970
    },
    {
      "epoch": 8.628318584070797,
      "grad_norm": 0.020375702530145645,
      "learning_rate": 2.8470576363586633e-06,
      "loss": 0.0024,
      "num_input_tokens_seen": 273824,
      "step": 975
    },
    {
      "epoch": 8.672566371681416,
      "grad_norm": 0.15513527393341064,
      "learning_rate": 2.67074896531056e-06,
      "loss": 0.0197,
      "num_input_tokens_seen": 275024,
      "step": 980
    },
    {
      "epoch": 8.716814159292035,
      "grad_norm": 0.011211968027055264,
      "learning_rate": 2.499767057216193e-06,
      "loss": 0.0002,
      "num_input_tokens_seen": 276256,
      "step": 985
    },
    {
      "epoch": 8.761061946902656,
      "grad_norm": 0.005273541435599327,
      "learning_rate": 2.334152700729511e-06,
      "loss": 0.0003,
      "num_input_tokens_seen": 277536,
      "step": 990
    },
    {
      "epoch": 8.805309734513274,
      "grad_norm": 0.009691768325865269,
      "learning_rate": 2.1739454040461794e-06,
      "loss": 0.0004,
      "num_input_tokens_seen": 279104,
      "step": 995
    },
    {
      "epoch": 8.849557522123893,
      "grad_norm": 0.2336866408586502,
      "learning_rate": 2.019183385478654e-06,
      "loss": 0.001,
      "num_input_tokens_seen": 280480,
      "step": 1000
    },
    {
      "epoch": 8.893805309734514,
      "grad_norm": 1.064756155014038,
      "learning_rate": 1.8699035643389928e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 281664,
      "step": 1005
    },
    {
      "epoch": 8.938053097345133,
      "grad_norm": 0.006707900669425726,
      "learning_rate": 1.7261415521315798e-06,
      "loss": 0.0119,
      "num_input_tokens_seen": 283248,
      "step": 1010
    },
    {
      "epoch": 8.982300884955752,
      "grad_norm": 3.824517250061035,
      "learning_rate": 1.587931644057783e-06,
      "loss": 0.0056,
      "num_input_tokens_seen": 284400,
      "step": 1015
    },
    {
      "epoch": 9.026548672566372,
      "grad_norm": 0.0035742255859076977,
      "learning_rate": 1.455306810834678e-06,
      "loss": 0.012,
      "num_input_tokens_seen": 285880,
      "step": 1020
    },
    {
      "epoch": 9.070796460176991,
      "grad_norm": 0.04793722555041313,
      "learning_rate": 1.3282986908296713e-06,
      "loss": 0.0057,
      "num_input_tokens_seen": 287480,
      "step": 1025
    },
    {
      "epoch": 9.079646017699115,
      "eval_loss": 0.41097939014434814,
      "eval_runtime": 0.7481,
      "eval_samples_per_second": 33.417,
      "eval_steps_per_second": 17.377,
      "num_input_tokens_seen": 287928,
      "step": 1026
    },
    {
      "epoch": 9.11504424778761,
      "grad_norm": 0.022159403190016747,
      "learning_rate": 1.2069375825130292e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 289144,
      "step": 1030
    },
    {
      "epoch": 9.15929203539823,
      "grad_norm": 0.18064860999584198,
      "learning_rate": 1.091252437230003e-06,
      "loss": 0.006,
      "num_input_tokens_seen": 290504,
      "step": 1035
    },
    {
      "epoch": 9.20353982300885,
      "grad_norm": 0.00854189321398735,
      "learning_rate": 9.812708522943164e-07,
      "loss": 0.0035,
      "num_input_tokens_seen": 291784,
      "step": 1040
    },
    {
      "epoch": 9.247787610619469,
      "grad_norm": 0.007520325947552919,
      "learning_rate": 8.770190644047077e-07,
      "loss": 0.0125,
      "num_input_tokens_seen": 292904,
      "step": 1045
    },
    {
      "epoch": 9.29203539823009,
      "grad_norm": 0.004971267655491829,
      "learning_rate": 7.785219433859847e-07,
      "loss": 0.0004,
      "num_input_tokens_seen": 294360,
      "step": 1050
    },
    {
      "epoch": 9.336283185840708,
      "grad_norm": 0.011202730238437653,
      "learning_rate": 6.858029862562343e-07,
      "loss": 0.0002,
      "num_input_tokens_seen": 295992,
      "step": 1055
    },
    {
      "epoch": 9.380530973451327,
      "grad_norm": 0.04648682847619057,
      "learning_rate": 5.988843116214616e-07,
      "loss": 0.0005,
      "num_input_tokens_seen": 297272,
      "step": 1060
    },
    {
      "epoch": 9.424778761061948,
      "grad_norm": 0.12177152931690216,
      "learning_rate": 5.177866543990689e-07,
      "loss": 0.0005,
      "num_input_tokens_seen": 298360,
      "step": 1065
    },
    {
      "epoch": 9.469026548672566,
      "grad_norm": 0.02386440522968769,
      "learning_rate": 4.425293608714548e-07,
      "loss": 0.0002,
      "num_input_tokens_seen": 299656,
      "step": 1070
    },
    {
      "epoch": 9.513274336283185,
      "grad_norm": 0.010457984171807766,
      "learning_rate": 3.7313038407083003e-07,
      "loss": 0.0003,
      "num_input_tokens_seen": 301432,
      "step": 1075
    },
    {
      "epoch": 9.557522123893806,
      "grad_norm": 0.008816664107143879,
      "learning_rate": 3.0960627949644107e-07,
      "loss": 0.0004,
      "num_input_tokens_seen": 302952,
      "step": 1080
    },
    {
      "epoch": 9.584070796460177,
      "eval_loss": 0.3926541209220886,
      "eval_runtime": 0.7493,
      "eval_samples_per_second": 33.363,
      "eval_steps_per_second": 17.349,
      "num_input_tokens_seen": 303800,
      "step": 1083
    },
    {
      "epoch": 9.601769911504425,
      "grad_norm": 0.08803759515285492,
      "learning_rate": 2.5197220116515987e-07,
      "loss": 0.0007,
      "num_input_tokens_seen": 304296,
      "step": 1085
    },
    {
      "epoch": 9.646017699115044,
      "grad_norm": 0.0047805532813072205,
      "learning_rate": 2.0024189799641435e-07,
      "loss": 0.0007,
      "num_input_tokens_seen": 305784,
      "step": 1090
    },
    {
      "epoch": 9.690265486725664,
      "grad_norm": 0.006290826015174389,
      "learning_rate": 1.5442771053230665e-07,
      "loss": 0.0003,
      "num_input_tokens_seen": 307592,
      "step": 1095
    },
    {
      "epoch": 9.734513274336283,
      "grad_norm": 0.010769442655146122,
      "learning_rate": 1.1454056799370672e-07,
      "loss": 0.0025,
      "num_input_tokens_seen": 308760,
      "step": 1100
    },
    {
      "epoch": 9.778761061946902,
      "grad_norm": 0.006454398389905691,
      "learning_rate": 8.058998567304633e-08,
      "loss": 0.0004,
      "num_input_tokens_seen": 310040,
      "step": 1105
    },
    {
      "epoch": 9.823008849557523,
      "grad_norm": 0.004815297666937113,
      "learning_rate": 5.2584062664368105e-08,
      "loss": 0.0004,
      "num_input_tokens_seen": 311752,
      "step": 1110
    },
    {
      "epoch": 9.867256637168142,
      "grad_norm": 0.0038188346661627293,
      "learning_rate": 3.0529479931265535e-08,
      "loss": 0.0003,
      "num_input_tokens_seen": 313128,
      "step": 1115
    },
    {
      "epoch": 9.91150442477876,
      "grad_norm": 0.00405368534848094,
      "learning_rate": 1.443149871309113e-08,
      "loss": 0.0008,
      "num_input_tokens_seen": 314568,
      "step": 1120
    },
    {
      "epoch": 9.955752212389381,
      "grad_norm": 0.009060665965080261,
      "learning_rate": 4.293959269863201e-09,
      "loss": 0.0002,
      "num_input_tokens_seen": 315800,
      "step": 1125
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.004397038370370865,
      "learning_rate": 1.192799661514643e-10,
      "loss": 0.0826,
      "num_input_tokens_seen": 316840,
      "step": 1130
    },
    {
      "epoch": 10.0,
      "num_input_tokens_seen": 316840,
      "step": 1130,
      "total_flos": 1.426716199845888e+16,
      "train_loss": 0.3219029305349784,
      "train_runtime": 156.6102,
      "train_samples_per_second": 14.367,
      "train_steps_per_second": 7.215
    }
  ],
  "logging_steps": 5,
  "max_steps": 1130,
  "num_input_tokens_seen": 316840,
  "num_train_epochs": 10,
  "save_steps": 57,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.426716199845888e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}