{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.5384615384615383,
  "eval_steps": 10,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003076923076923077,
      "eval_loss": 1.4241737127304077,
      "eval_runtime": 2.2418,
      "eval_samples_per_second": 50.852,
      "eval_steps_per_second": 2.676,
      "step": 2
    },
    {
      "epoch": 0.015384615384615385,
      "grad_norm": 11.650300979614258,
      "learning_rate": 4.395604395604396e-07,
      "loss": 1.7785,
      "step": 10
    },
    {
      "epoch": 0.015384615384615385,
      "eval_loss": 1.4170304536819458,
      "eval_runtime": 2.0399,
      "eval_samples_per_second": 55.885,
      "eval_steps_per_second": 2.941,
      "step": 10
    },
    {
      "epoch": 0.03076923076923077,
      "grad_norm": 14.07431697845459,
      "learning_rate": 8.791208791208792e-07,
      "loss": 1.4133,
      "step": 20
    },
    {
      "epoch": 0.03076923076923077,
      "eval_loss": 1.32579505443573,
      "eval_runtime": 2.0523,
      "eval_samples_per_second": 55.546,
      "eval_steps_per_second": 2.923,
      "step": 20
    },
    {
      "epoch": 0.046153846153846156,
      "grad_norm": 10.966695785522461,
      "learning_rate": 1.3186813186813187e-06,
      "loss": 1.0466,
      "step": 30
    },
    {
      "epoch": 0.046153846153846156,
      "eval_loss": 0.9353072643280029,
      "eval_runtime": 2.6297,
      "eval_samples_per_second": 43.351,
      "eval_steps_per_second": 2.282,
      "step": 30
    },
    {
      "epoch": 0.06153846153846154,
      "grad_norm": 9.173104286193848,
      "learning_rate": 1.7582417582417585e-06,
      "loss": 0.8387,
      "step": 40
    },
    {
      "epoch": 0.06153846153846154,
      "eval_loss": 0.8081339597702026,
      "eval_runtime": 2.1542,
      "eval_samples_per_second": 52.919,
      "eval_steps_per_second": 2.785,
      "step": 40
    },
    {
      "epoch": 0.07692307692307693,
      "grad_norm": 8.724108695983887,
      "learning_rate": 2.197802197802198e-06,
      "loss": 0.6723,
      "step": 50
    },
    {
      "epoch": 0.07692307692307693,
      "eval_loss": 0.7670798301696777,
      "eval_runtime": 2.0511,
      "eval_samples_per_second": 55.58,
      "eval_steps_per_second": 2.925,
      "step": 50
    },
    {
      "epoch": 0.09230769230769231,
      "grad_norm": 30.25709342956543,
      "learning_rate": 2.6373626373626375e-06,
      "loss": 0.7863,
      "step": 60
    },
    {
      "epoch": 0.09230769230769231,
      "eval_loss": 0.750013530254364,
      "eval_runtime": 2.0411,
      "eval_samples_per_second": 55.851,
      "eval_steps_per_second": 2.94,
      "step": 60
    },
    {
      "epoch": 0.1076923076923077,
      "grad_norm": 26.09903335571289,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.7312,
      "step": 70
    },
    {
      "epoch": 0.1076923076923077,
      "eval_loss": 0.7440299987792969,
      "eval_runtime": 2.0505,
      "eval_samples_per_second": 55.595,
      "eval_steps_per_second": 2.926,
      "step": 70
    },
    {
      "epoch": 0.12307692307692308,
      "grad_norm": 13.608738899230957,
      "learning_rate": 3.516483516483517e-06,
      "loss": 0.7022,
      "step": 80
    },
    {
      "epoch": 0.12307692307692308,
      "eval_loss": 0.7316409945487976,
      "eval_runtime": 2.0448,
      "eval_samples_per_second": 55.75,
      "eval_steps_per_second": 2.934,
      "step": 80
    },
    {
      "epoch": 0.13846153846153847,
      "grad_norm": 12.300930976867676,
      "learning_rate": 3.9560439560439565e-06,
      "loss": 0.6959,
      "step": 90
    },
    {
      "epoch": 0.13846153846153847,
      "eval_loss": 0.7208518981933594,
      "eval_runtime": 2.0423,
      "eval_samples_per_second": 55.819,
      "eval_steps_per_second": 2.938,
      "step": 90
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 10.648837089538574,
      "learning_rate": 4.395604395604396e-06,
      "loss": 0.7368,
      "step": 100
    },
    {
      "epoch": 0.15384615384615385,
      "eval_loss": 0.7109653353691101,
      "eval_runtime": 2.0386,
      "eval_samples_per_second": 55.921,
      "eval_steps_per_second": 2.943,
      "step": 100
    },
    {
      "epoch": 0.16923076923076924,
      "grad_norm": 14.025050163269043,
      "learning_rate": 4.8351648351648355e-06,
      "loss": 0.6095,
      "step": 110
    },
    {
      "epoch": 0.16923076923076924,
      "eval_loss": 0.7375777959823608,
      "eval_runtime": 2.0429,
      "eval_samples_per_second": 55.802,
      "eval_steps_per_second": 2.937,
      "step": 110
    },
    {
      "epoch": 0.18461538461538463,
      "grad_norm": 11.35698413848877,
      "learning_rate": 5.274725274725275e-06,
      "loss": 0.5998,
      "step": 120
    },
    {
      "epoch": 0.18461538461538463,
      "eval_loss": 0.7044270038604736,
      "eval_runtime": 2.0444,
      "eval_samples_per_second": 55.763,
      "eval_steps_per_second": 2.935,
      "step": 120
    },
    {
      "epoch": 0.2,
      "grad_norm": 6.319034099578857,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 0.5771,
      "step": 130
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.7100598812103271,
      "eval_runtime": 2.0372,
      "eval_samples_per_second": 55.958,
      "eval_steps_per_second": 2.945,
      "step": 130
    },
    {
      "epoch": 0.2153846153846154,
      "grad_norm": 2.982799530029297,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.6435,
      "step": 140
    },
    {
      "epoch": 0.2153846153846154,
      "eval_loss": 0.7007233500480652,
      "eval_runtime": 2.0408,
      "eval_samples_per_second": 55.859,
      "eval_steps_per_second": 2.94,
      "step": 140
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 8.233396530151367,
      "learning_rate": 6.5934065934065935e-06,
      "loss": 0.6292,
      "step": 150
    },
    {
      "epoch": 0.23076923076923078,
      "eval_loss": 0.7087807655334473,
      "eval_runtime": 2.0373,
      "eval_samples_per_second": 55.956,
      "eval_steps_per_second": 2.945,
      "step": 150
    },
    {
      "epoch": 0.24615384615384617,
      "grad_norm": 2.763383388519287,
      "learning_rate": 7.032967032967034e-06,
      "loss": 0.6035,
      "step": 160
    },
    {
      "epoch": 0.24615384615384617,
      "eval_loss": 0.7013908624649048,
      "eval_runtime": 2.043,
      "eval_samples_per_second": 55.8,
      "eval_steps_per_second": 2.937,
      "step": 160
    },
    {
      "epoch": 0.26153846153846155,
      "grad_norm": 2.742600679397583,
      "learning_rate": 7.472527472527473e-06,
      "loss": 0.6948,
      "step": 170
    },
    {
      "epoch": 0.26153846153846155,
      "eval_loss": 0.7081724405288696,
      "eval_runtime": 2.0395,
      "eval_samples_per_second": 55.896,
      "eval_steps_per_second": 2.942,
      "step": 170
    },
    {
      "epoch": 0.27692307692307694,
      "grad_norm": 6.114007949829102,
      "learning_rate": 7.912087912087913e-06,
      "loss": 0.6616,
      "step": 180
    },
    {
      "epoch": 0.27692307692307694,
      "eval_loss": 0.700532853603363,
      "eval_runtime": 2.0445,
      "eval_samples_per_second": 55.76,
      "eval_steps_per_second": 2.935,
      "step": 180
    },
    {
      "epoch": 0.2923076923076923,
      "grad_norm": 3.872872829437256,
      "learning_rate": 8.351648351648353e-06,
      "loss": 0.5296,
      "step": 190
    },
    {
      "epoch": 0.2923076923076923,
      "eval_loss": 0.7047591209411621,
      "eval_runtime": 2.0453,
      "eval_samples_per_second": 55.737,
      "eval_steps_per_second": 2.934,
      "step": 190
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 4.802947044372559,
      "learning_rate": 8.791208791208792e-06,
      "loss": 0.6539,
      "step": 200
    },
    {
      "epoch": 0.3076923076923077,
      "eval_loss": 0.705501914024353,
      "eval_runtime": 2.0419,
      "eval_samples_per_second": 55.829,
      "eval_steps_per_second": 2.938,
      "step": 200
    },
    {
      "epoch": 0.3230769230769231,
      "grad_norm": 3.390991687774658,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.6817,
      "step": 210
    },
    {
      "epoch": 0.3230769230769231,
      "eval_loss": 0.7018715739250183,
      "eval_runtime": 2.0542,
      "eval_samples_per_second": 55.495,
      "eval_steps_per_second": 2.921,
      "step": 210
    },
    {
      "epoch": 0.3384615384615385,
      "grad_norm": 6.308581829071045,
      "learning_rate": 9.670329670329671e-06,
      "loss": 0.6024,
      "step": 220
    },
    {
      "epoch": 0.3384615384615385,
      "eval_loss": 0.7041005492210388,
      "eval_runtime": 2.0378,
      "eval_samples_per_second": 55.943,
      "eval_steps_per_second": 2.944,
      "step": 220
    },
    {
      "epoch": 0.35384615384615387,
      "grad_norm": 2.0694475173950195,
      "learning_rate": 1.010989010989011e-05,
      "loss": 0.5431,
      "step": 230
    },
    {
      "epoch": 0.35384615384615387,
      "eval_loss": 0.7189098000526428,
      "eval_runtime": 2.0403,
      "eval_samples_per_second": 55.873,
      "eval_steps_per_second": 2.941,
      "step": 230
    },
    {
      "epoch": 0.36923076923076925,
      "grad_norm": 2.511876106262207,
      "learning_rate": 1.054945054945055e-05,
      "loss": 0.7137,
      "step": 240
    },
    {
      "epoch": 0.36923076923076925,
      "eval_loss": 0.7117995023727417,
      "eval_runtime": 2.0312,
      "eval_samples_per_second": 56.125,
      "eval_steps_per_second": 2.954,
      "step": 240
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 3.0961618423461914,
      "learning_rate": 1.098901098901099e-05,
      "loss": 0.7315,
      "step": 250
    },
    {
      "epoch": 0.38461538461538464,
      "eval_loss": 0.7184381484985352,
      "eval_runtime": 2.0407,
      "eval_samples_per_second": 55.862,
      "eval_steps_per_second": 2.94,
      "step": 250
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.529021978378296,
      "learning_rate": 1.1428571428571429e-05,
      "loss": 0.6812,
      "step": 260
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.7177315950393677,
      "eval_runtime": 2.039,
      "eval_samples_per_second": 55.91,
      "eval_steps_per_second": 2.943,
      "step": 260
    },
    {
      "epoch": 0.4153846153846154,
      "grad_norm": 2.3414862155914307,
      "learning_rate": 1.186813186813187e-05,
      "loss": 0.6943,
      "step": 270
    },
    {
      "epoch": 0.4153846153846154,
      "eval_loss": 0.721341073513031,
      "eval_runtime": 2.052,
      "eval_samples_per_second": 55.555,
      "eval_steps_per_second": 2.924,
      "step": 270
    },
    {
      "epoch": 0.4307692307692308,
      "grad_norm": 4.085804462432861,
      "learning_rate": 1.230769230769231e-05,
      "loss": 0.7691,
      "step": 280
    },
    {
      "epoch": 0.4307692307692308,
      "eval_loss": 0.7427098751068115,
      "eval_runtime": 2.0471,
      "eval_samples_per_second": 55.689,
      "eval_steps_per_second": 2.931,
      "step": 280
    },
    {
      "epoch": 0.4461538461538462,
      "grad_norm": 3.791297197341919,
      "learning_rate": 1.2747252747252747e-05,
      "loss": 0.5522,
      "step": 290
    },
    {
      "epoch": 0.4461538461538462,
      "eval_loss": 0.7322326302528381,
      "eval_runtime": 2.0475,
      "eval_samples_per_second": 55.678,
      "eval_steps_per_second": 2.93,
      "step": 290
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 4.031556606292725,
      "learning_rate": 1.3186813186813187e-05,
      "loss": 0.6837,
      "step": 300
    },
    {
      "epoch": 0.46153846153846156,
      "eval_loss": 0.7213509678840637,
      "eval_runtime": 2.0384,
      "eval_samples_per_second": 55.925,
      "eval_steps_per_second": 2.943,
      "step": 300
    },
    {
      "epoch": 0.47692307692307695,
      "grad_norm": 2.4052727222442627,
      "learning_rate": 1.3626373626373627e-05,
      "loss": 0.587,
      "step": 310
    },
    {
      "epoch": 0.47692307692307695,
      "eval_loss": 0.7282149791717529,
      "eval_runtime": 2.0446,
      "eval_samples_per_second": 55.758,
      "eval_steps_per_second": 2.935,
      "step": 310
    },
    {
      "epoch": 0.49230769230769234,
      "grad_norm": 3.849602699279785,
      "learning_rate": 1.4065934065934068e-05,
      "loss": 0.6275,
      "step": 320
    },
    {
      "epoch": 0.49230769230769234,
      "eval_loss": 0.7264513373374939,
      "eval_runtime": 2.0589,
      "eval_samples_per_second": 55.369,
      "eval_steps_per_second": 2.914,
      "step": 320
    },
    {
      "epoch": 0.5076923076923077,
      "grad_norm": 2.8470089435577393,
      "learning_rate": 1.4505494505494506e-05,
      "loss": 0.7207,
      "step": 330
    },
    {
      "epoch": 0.5076923076923077,
      "eval_loss": 0.7379302978515625,
      "eval_runtime": 2.0433,
      "eval_samples_per_second": 55.793,
      "eval_steps_per_second": 2.936,
      "step": 330
    },
    {
      "epoch": 0.5230769230769231,
      "grad_norm": 2.827406167984009,
      "learning_rate": 1.4945054945054947e-05,
      "loss": 0.7876,
      "step": 340
    },
    {
      "epoch": 0.5230769230769231,
      "eval_loss": 0.7334662079811096,
      "eval_runtime": 2.0482,
      "eval_samples_per_second": 55.658,
      "eval_steps_per_second": 2.929,
      "step": 340
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 3.079512119293213,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.7344,
      "step": 350
    },
    {
      "epoch": 0.5384615384615384,
      "eval_loss": 0.7296246886253357,
      "eval_runtime": 2.0432,
      "eval_samples_per_second": 55.795,
      "eval_steps_per_second": 2.937,
      "step": 350
    },
    {
      "epoch": 0.5538461538461539,
      "grad_norm": 3.95023512840271,
      "learning_rate": 1.5824175824175826e-05,
      "loss": 0.7414,
      "step": 360
    },
    {
      "epoch": 0.5538461538461539,
      "eval_loss": 0.7329455018043518,
      "eval_runtime": 2.0374,
      "eval_samples_per_second": 55.954,
      "eval_steps_per_second": 2.945,
      "step": 360
    },
    {
      "epoch": 0.5692307692307692,
      "grad_norm": 2.8334782123565674,
      "learning_rate": 1.6263736263736265e-05,
      "loss": 0.5957,
      "step": 370
    },
    {
      "epoch": 0.5692307692307692,
      "eval_loss": 0.7440397143363953,
      "eval_runtime": 2.0406,
      "eval_samples_per_second": 55.866,
      "eval_steps_per_second": 2.94,
      "step": 370
    },
    {
      "epoch": 0.5846153846153846,
      "grad_norm": 11.327688217163086,
      "learning_rate": 1.6703296703296707e-05,
      "loss": 0.714,
      "step": 380
    },
    {
      "epoch": 0.5846153846153846,
      "eval_loss": 0.7371141910552979,
      "eval_runtime": 2.0417,
      "eval_samples_per_second": 55.834,
      "eval_steps_per_second": 2.939,
      "step": 380
    },
    {
      "epoch": 0.6,
      "grad_norm": 2.3651175498962402,
      "learning_rate": 1.7142857142857142e-05,
      "loss": 0.6974,
      "step": 390
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.7449064254760742,
      "eval_runtime": 2.0515,
      "eval_samples_per_second": 55.569,
      "eval_steps_per_second": 2.925,
      "step": 390
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 3.6230828762054443,
      "learning_rate": 1.7582417582417584e-05,
      "loss": 0.6405,
      "step": 400
    },
    {
      "epoch": 0.6153846153846154,
      "eval_loss": 0.7573131322860718,
      "eval_runtime": 2.0602,
      "eval_samples_per_second": 55.335,
      "eval_steps_per_second": 2.912,
      "step": 400
    },
    {
      "epoch": 0.6307692307692307,
      "grad_norm": 4.608823299407959,
      "learning_rate": 1.8021978021978023e-05,
      "loss": 0.675,
      "step": 410
    },
    {
      "epoch": 0.6307692307692307,
      "eval_loss": 0.7619711756706238,
      "eval_runtime": 2.0552,
      "eval_samples_per_second": 55.468,
      "eval_steps_per_second": 2.919,
      "step": 410
    },
    {
      "epoch": 0.6461538461538462,
      "grad_norm": 2.838009834289551,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.6799,
      "step": 420
    },
    {
      "epoch": 0.6461538461538462,
      "eval_loss": 0.7572401165962219,
      "eval_runtime": 2.0487,
      "eval_samples_per_second": 55.645,
      "eval_steps_per_second": 2.929,
      "step": 420
    },
    {
      "epoch": 0.6615384615384615,
      "grad_norm": 3.080890655517578,
      "learning_rate": 1.8901098901098903e-05,
      "loss": 0.657,
      "step": 430
    },
    {
      "epoch": 0.6615384615384615,
      "eval_loss": 0.749286413192749,
      "eval_runtime": 2.0418,
      "eval_samples_per_second": 55.832,
      "eval_steps_per_second": 2.939,
      "step": 430
    },
    {
      "epoch": 0.676923076923077,
      "grad_norm": 2.4955101013183594,
      "learning_rate": 1.9340659340659342e-05,
      "loss": 0.5658,
      "step": 440
    },
    {
      "epoch": 0.676923076923077,
      "eval_loss": 0.752851128578186,
      "eval_runtime": 2.0515,
      "eval_samples_per_second": 55.568,
      "eval_steps_per_second": 2.925,
      "step": 440
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 3.3726329803466797,
      "learning_rate": 1.9780219780219784e-05,
      "loss": 0.5746,
      "step": 450
    },
    {
      "epoch": 0.6923076923076923,
      "eval_loss": 0.7596186399459839,
      "eval_runtime": 2.0458,
      "eval_samples_per_second": 55.723,
      "eval_steps_per_second": 2.933,
      "step": 450
    },
    {
      "epoch": 0.7076923076923077,
      "grad_norm": 9.614293098449707,
      "learning_rate": 1.9999926429888597e-05,
      "loss": 0.6832,
      "step": 460
    },
    {
      "epoch": 0.7076923076923077,
      "eval_loss": 0.7571460604667664,
      "eval_runtime": 2.042,
      "eval_samples_per_second": 55.827,
      "eval_steps_per_second": 2.938,
      "step": 460
    },
    {
      "epoch": 0.7230769230769231,
      "grad_norm": 2.3883635997772217,
      "learning_rate": 1.9999337875492412e-05,
      "loss": 0.7594,
      "step": 470
    },
    {
      "epoch": 0.7230769230769231,
      "eval_loss": 0.7611674666404724,
      "eval_runtime": 2.0388,
      "eval_samples_per_second": 55.915,
      "eval_steps_per_second": 2.943,
      "step": 470
    },
    {
      "epoch": 0.7384615384615385,
      "grad_norm": 5.172886848449707,
      "learning_rate": 1.999816080133992e-05,
      "loss": 0.7668,
      "step": 480
    },
    {
      "epoch": 0.7384615384615385,
      "eval_loss": 0.7736306190490723,
      "eval_runtime": 2.0996,
      "eval_samples_per_second": 54.297,
      "eval_steps_per_second": 2.858,
      "step": 480
    },
    {
      "epoch": 0.7538461538461538,
      "grad_norm": 3.6946752071380615,
      "learning_rate": 1.9996395276708856e-05,
      "loss": 0.749,
      "step": 490
    },
    {
      "epoch": 0.7538461538461538,
      "eval_loss": 0.7762741446495056,
      "eval_runtime": 2.0394,
      "eval_samples_per_second": 55.899,
      "eval_steps_per_second": 2.942,
      "step": 490
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 1.411451816558838,
      "learning_rate": 1.9994041405510705e-05,
      "loss": 0.6495,
      "step": 500
    },
    {
      "epoch": 0.7692307692307693,
      "eval_loss": 0.7672939896583557,
      "eval_runtime": 2.0418,
      "eval_samples_per_second": 55.833,
      "eval_steps_per_second": 2.939,
      "step": 500
    },
    {
      "epoch": 0.7846153846153846,
      "grad_norm": 3.8847391605377197,
      "learning_rate": 1.9991099326284616e-05,
      "loss": 0.7146,
      "step": 510
    },
    {
      "epoch": 0.7846153846153846,
      "eval_loss": 0.7738469839096069,
      "eval_runtime": 2.0369,
      "eval_samples_per_second": 55.967,
      "eval_steps_per_second": 2.946,
      "step": 510
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.122893214225769,
      "learning_rate": 1.9987569212189224e-05,
      "loss": 0.7931,
      "step": 520
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.7790060043334961,
      "eval_runtime": 2.0674,
      "eval_samples_per_second": 55.143,
      "eval_steps_per_second": 2.902,
      "step": 520
    },
    {
      "epoch": 0.8153846153846154,
      "grad_norm": 3.256478786468506,
      "learning_rate": 1.998345127099248e-05,
      "loss": 0.644,
      "step": 530
    },
    {
      "epoch": 0.8153846153846154,
      "eval_loss": 0.7947230935096741,
      "eval_runtime": 2.0478,
      "eval_samples_per_second": 55.67,
      "eval_steps_per_second": 2.93,
      "step": 530
    },
    {
      "epoch": 0.8307692307692308,
      "grad_norm": 3.0072100162506104,
      "learning_rate": 1.99787457450594e-05,
      "loss": 0.7743,
      "step": 540
    },
    {
      "epoch": 0.8307692307692308,
      "eval_loss": 0.7725875377655029,
      "eval_runtime": 2.0492,
      "eval_samples_per_second": 55.632,
      "eval_steps_per_second": 2.928,
      "step": 540
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 6.6161112785339355,
      "learning_rate": 1.997345291133783e-05,
      "loss": 0.9146,
      "step": 550
    },
    {
      "epoch": 0.8461538461538461,
      "eval_loss": 0.7745257019996643,
      "eval_runtime": 2.0485,
      "eval_samples_per_second": 55.652,
      "eval_steps_per_second": 2.929,
      "step": 550
    },
    {
      "epoch": 0.8615384615384616,
      "grad_norm": 17.896169662475586,
      "learning_rate": 1.9967573081342103e-05,
      "loss": 0.7511,
      "step": 560
    },
    {
      "epoch": 0.8615384615384616,
      "eval_loss": 0.7743226289749146,
      "eval_runtime": 2.0433,
      "eval_samples_per_second": 55.793,
      "eval_steps_per_second": 2.936,
      "step": 560
    },
    {
      "epoch": 0.8769230769230769,
      "grad_norm": 8.162055969238281,
      "learning_rate": 1.996110660113475e-05,
      "loss": 0.6427,
      "step": 570
    },
    {
      "epoch": 0.8769230769230769,
      "eval_loss": 0.7627045512199402,
      "eval_runtime": 2.0492,
      "eval_samples_per_second": 55.632,
      "eval_steps_per_second": 2.928,
      "step": 570
    },
    {
      "epoch": 0.8923076923076924,
      "grad_norm": 2.4918031692504883,
      "learning_rate": 1.995405385130611e-05,
      "loss": 0.7831,
      "step": 580
    },
    {
      "epoch": 0.8923076923076924,
      "eval_loss": 0.7645999789237976,
      "eval_runtime": 2.0464,
      "eval_samples_per_second": 55.707,
      "eval_steps_per_second": 2.932,
      "step": 580
    },
    {
      "epoch": 0.9076923076923077,
      "grad_norm": 2.0997047424316406,
      "learning_rate": 1.9946415246951928e-05,
      "loss": 0.7018,
      "step": 590
    },
    {
      "epoch": 0.9076923076923077,
      "eval_loss": 0.7743611931800842,
      "eval_runtime": 2.0438,
      "eval_samples_per_second": 55.779,
      "eval_steps_per_second": 2.936,
      "step": 590
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 3.648905038833618,
      "learning_rate": 1.9938191237648924e-05,
      "loss": 0.8711,
      "step": 600
    },
    {
      "epoch": 0.9230769230769231,
      "eval_loss": 0.7733822464942932,
      "eval_runtime": 2.0548,
      "eval_samples_per_second": 55.481,
      "eval_steps_per_second": 2.92,
      "step": 600
    },
    {
      "epoch": 0.9384615384615385,
      "grad_norm": 1.8373923301696777,
      "learning_rate": 1.992938230742835e-05,
      "loss": 0.7424,
      "step": 610
    },
    {
      "epoch": 0.9384615384615385,
      "eval_loss": 0.7707686424255371,
      "eval_runtime": 2.0497,
      "eval_samples_per_second": 55.619,
      "eval_steps_per_second": 2.927,
      "step": 610
    },
    {
      "epoch": 0.9538461538461539,
      "grad_norm": 2.7513844966888428,
      "learning_rate": 1.9919988974747473e-05,
      "loss": 0.7595,
      "step": 620
    },
    {
      "epoch": 0.9538461538461539,
      "eval_loss": 0.7717193961143494,
      "eval_runtime": 2.0354,
      "eval_samples_per_second": 56.009,
      "eval_steps_per_second": 2.948,
      "step": 620
    },
    {
      "epoch": 0.9692307692307692,
      "grad_norm": 3.6671864986419678,
      "learning_rate": 1.9910011792459086e-05,
      "loss": 0.7302,
      "step": 630
    },
    {
      "epoch": 0.9692307692307692,
      "eval_loss": 0.7755113840103149,
      "eval_runtime": 2.0422,
      "eval_samples_per_second": 55.822,
      "eval_steps_per_second": 2.938,
      "step": 630
    },
    {
      "epoch": 0.9846153846153847,
      "grad_norm": 4.106902122497559,
      "learning_rate": 1.9899451347778962e-05,
      "loss": 0.6058,
      "step": 640
    },
    {
      "epoch": 0.9846153846153847,
      "eval_loss": 0.7776271104812622,
      "eval_runtime": 2.0327,
      "eval_samples_per_second": 56.082,
      "eval_steps_per_second": 2.952,
      "step": 640
    },
    {
      "epoch": 1.0,
      "grad_norm": 3.700509786605835,
      "learning_rate": 1.9888308262251286e-05,
      "loss": 0.6814,
      "step": 650
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.778499960899353,
      "eval_runtime": 2.0341,
      "eval_samples_per_second": 56.044,
      "eval_steps_per_second": 2.95,
      "step": 650
    },
    {
      "epoch": 1.0153846153846153,
      "grad_norm": 2.705697536468506,
      "learning_rate": 1.9876583191712083e-05,
      "loss": 0.3862,
      "step": 660
    },
    {
      "epoch": 1.0153846153846153,
      "eval_loss": 0.8296825289726257,
      "eval_runtime": 2.0421,
      "eval_samples_per_second": 55.824,
      "eval_steps_per_second": 2.938,
      "step": 660
    },
    {
      "epoch": 1.0307692307692307,
      "grad_norm": 2.9704339504241943,
      "learning_rate": 1.9864276826250608e-05,
      "loss": 0.3792,
      "step": 670
    },
    {
      "epoch": 1.0307692307692307,
      "eval_loss": 0.8110206127166748,
      "eval_runtime": 2.0363,
      "eval_samples_per_second": 55.983,
      "eval_steps_per_second": 2.946,
      "step": 670
    },
    {
      "epoch": 1.0461538461538462,
      "grad_norm": 3.3613340854644775,
      "learning_rate": 1.9851389890168738e-05,
      "loss": 0.4167,
      "step": 680
    },
    {
      "epoch": 1.0461538461538462,
      "eval_loss": 0.8016878366470337,
      "eval_runtime": 2.0429,
      "eval_samples_per_second": 55.802,
      "eval_steps_per_second": 2.937,
      "step": 680
    },
    {
      "epoch": 1.0615384615384615,
      "grad_norm": 3.673887014389038,
      "learning_rate": 1.983792314193835e-05,
      "loss": 0.3572,
      "step": 690
    },
    {
      "epoch": 1.0615384615384615,
      "eval_loss": 0.8120518922805786,
      "eval_runtime": 2.0316,
      "eval_samples_per_second": 56.114,
      "eval_steps_per_second": 2.953,
      "step": 690
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.5336493253707886,
      "learning_rate": 1.9823877374156647e-05,
      "loss": 0.3411,
      "step": 700
    },
    {
      "epoch": 1.0769230769230769,
      "eval_loss": 0.8034031391143799,
      "eval_runtime": 2.0347,
      "eval_samples_per_second": 56.029,
      "eval_steps_per_second": 2.949,
      "step": 700
    },
    {
      "epoch": 1.0923076923076924,
      "grad_norm": 3.4111616611480713,
      "learning_rate": 1.9809253413499565e-05,
      "loss": 0.445,
      "step": 710
    },
    {
      "epoch": 1.0923076923076924,
      "eval_loss": 0.7985233664512634,
      "eval_runtime": 2.0544,
      "eval_samples_per_second": 55.49,
      "eval_steps_per_second": 2.921,
      "step": 710
    },
    {
      "epoch": 1.1076923076923078,
      "grad_norm": 3.9824275970458984,
      "learning_rate": 1.979405212067306e-05,
      "loss": 0.3657,
      "step": 720
    },
    {
      "epoch": 1.1076923076923078,
      "eval_loss": 0.8141591548919678,
      "eval_runtime": 2.0508,
      "eval_samples_per_second": 55.588,
      "eval_steps_per_second": 2.926,
      "step": 720
    },
    {
      "epoch": 1.123076923076923,
      "grad_norm": 1.543548583984375,
      "learning_rate": 1.9778274390362488e-05,
      "loss": 0.362,
      "step": 730
    },
    {
      "epoch": 1.123076923076923,
      "eval_loss": 0.8183783888816833,
      "eval_runtime": 2.047,
      "eval_samples_per_second": 55.691,
      "eval_steps_per_second": 2.931,
      "step": 730
    },
    {
      "epoch": 1.1384615384615384,
      "grad_norm": 2.2039008140563965,
      "learning_rate": 1.9761921151179937e-05,
      "loss": 0.4064,
      "step": 740
    },
    {
      "epoch": 1.1384615384615384,
      "eval_loss": 0.8234056830406189,
      "eval_runtime": 2.0415,
      "eval_samples_per_second": 55.842,
      "eval_steps_per_second": 2.939,
      "step": 740
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 3.632371425628662,
      "learning_rate": 1.9744993365609563e-05,
      "loss": 0.4081,
      "step": 750
    },
    {
      "epoch": 1.1538461538461537,
      "eval_loss": 0.8307737112045288,
      "eval_runtime": 2.0452,
      "eval_samples_per_second": 55.74,
      "eval_steps_per_second": 2.934,
      "step": 750
    },
    {
      "epoch": 1.1692307692307693,
      "grad_norm": 1.8475432395935059,
      "learning_rate": 1.9727492029950965e-05,
      "loss": 0.3464,
      "step": 760
    },
    {
      "epoch": 1.1692307692307693,
      "eval_loss": 0.8189820647239685,
      "eval_runtime": 2.0369,
      "eval_samples_per_second": 55.968,
      "eval_steps_per_second": 2.946,
      "step": 760
    },
    {
      "epoch": 1.1846153846153846,
      "grad_norm": 2.644315242767334,
      "learning_rate": 1.9709418174260523e-05,
      "loss": 0.3052,
      "step": 770
    },
    {
      "epoch": 1.1846153846153846,
      "eval_loss": 0.8321068286895752,
      "eval_runtime": 2.0399,
      "eval_samples_per_second": 55.886,
      "eval_steps_per_second": 2.941,
      "step": 770
    },
    {
      "epoch": 1.2,
      "grad_norm": 3.449650764465332,
      "learning_rate": 1.969077286229078e-05,
      "loss": 0.3573,
      "step": 780
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.8303280472755432,
      "eval_runtime": 2.0434,
      "eval_samples_per_second": 55.789,
      "eval_steps_per_second": 2.936,
      "step": 780
    },
    {
      "epoch": 1.2153846153846155,
      "grad_norm": 3.0072035789489746,
      "learning_rate": 1.967155719142785e-05,
      "loss": 0.4151,
      "step": 790
    },
    {
      "epoch": 1.2153846153846155,
      "eval_loss": 0.8407182693481445,
      "eval_runtime": 2.0518,
      "eval_samples_per_second": 55.562,
      "eval_steps_per_second": 2.924,
      "step": 790
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 3.1887924671173096,
      "learning_rate": 1.9651772292626804e-05,
      "loss": 0.4201,
      "step": 800
    },
    {
      "epoch": 1.2307692307692308,
      "eval_loss": 0.8193632364273071,
      "eval_runtime": 2.0387,
      "eval_samples_per_second": 55.919,
      "eval_steps_per_second": 2.943,
      "step": 800
    },
    {
      "epoch": 1.2461538461538462,
      "grad_norm": 2.6908791065216064,
      "learning_rate": 1.9631419330345128e-05,
      "loss": 0.3639,
      "step": 810
    },
    {
      "epoch": 1.2461538461538462,
      "eval_loss": 0.8185432553291321,
      "eval_runtime": 2.0537,
      "eval_samples_per_second": 55.509,
      "eval_steps_per_second": 2.922,
      "step": 810
    },
    {
      "epoch": 1.2615384615384615,
      "grad_norm": 2.4561610221862793,
      "learning_rate": 1.961049950247418e-05,
      "loss": 0.4031,
      "step": 820
    },
    {
      "epoch": 1.2615384615384615,
      "eval_loss": 0.8187156915664673,
      "eval_runtime": 2.0342,
      "eval_samples_per_second": 56.041,
      "eval_steps_per_second": 2.95,
      "step": 820
    },
    {
      "epoch": 1.2769230769230768,
      "grad_norm": 3.1867499351501465,
      "learning_rate": 1.9589014040268678e-05,
      "loss": 0.4187,
      "step": 830
    },
    {
      "epoch": 1.2769230769230768,
      "eval_loss": 0.807949960231781,
      "eval_runtime": 2.0362,
      "eval_samples_per_second": 55.987,
      "eval_steps_per_second": 2.947,
      "step": 830
    },
    {
      "epoch": 1.2923076923076924,
      "grad_norm": 2.2672019004821777,
      "learning_rate": 1.9566964208274254e-05,
      "loss": 0.3999,
      "step": 840
    },
    {
      "epoch": 1.2923076923076924,
      "eval_loss": 0.8243932127952576,
      "eval_runtime": 2.0306,
      "eval_samples_per_second": 56.141,
      "eval_steps_per_second": 2.955,
      "step": 840
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.9958841800689697,
      "learning_rate": 1.954435130425301e-05,
      "loss": 0.3634,
      "step": 850
    },
    {
      "epoch": 1.3076923076923077,
      "eval_loss": 0.8296343088150024,
      "eval_runtime": 2.0474,
      "eval_samples_per_second": 55.681,
      "eval_steps_per_second": 2.931,
      "step": 850
    },
    {
      "epoch": 1.323076923076923,
      "grad_norm": 2.46744704246521,
      "learning_rate": 1.952117665910714e-05,
      "loss": 0.4121,
      "step": 860
    },
    {
      "epoch": 1.323076923076923,
      "eval_loss": 0.8178287744522095,
      "eval_runtime": 2.0413,
      "eval_samples_per_second": 55.847,
      "eval_steps_per_second": 2.939,
      "step": 860
    },
    {
      "epoch": 1.3384615384615386,
      "grad_norm": 3.1255440711975098,
      "learning_rate": 1.949744163680062e-05,
      "loss": 0.4283,
      "step": 870
    },
    {
      "epoch": 1.3384615384615386,
      "eval_loss": 0.8322147130966187,
      "eval_runtime": 2.0347,
      "eval_samples_per_second": 56.027,
      "eval_steps_per_second": 2.949,
      "step": 870
    },
    {
      "epoch": 1.353846153846154,
      "grad_norm": 3.667170763015747,
      "learning_rate": 1.9473147634278884e-05,
      "loss": 0.397,
      "step": 880
    },
    {
      "epoch": 1.353846153846154,
      "eval_loss": 0.821895956993103,
      "eval_runtime": 2.0641,
      "eval_samples_per_second": 55.23,
      "eval_steps_per_second": 2.907,
      "step": 880
    },
    {
      "epoch": 1.3692307692307693,
      "grad_norm": 1.867911696434021,
      "learning_rate": 1.9448296081386656e-05,
      "loss": 0.3877,
      "step": 890
    },
    {
      "epoch": 1.3692307692307693,
      "eval_loss": 0.8311596512794495,
      "eval_runtime": 2.0413,
      "eval_samples_per_second": 55.847,
      "eval_steps_per_second": 2.939,
      "step": 890
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 1.2144052982330322,
      "learning_rate": 1.9422888440783773e-05,
      "loss": 0.3512,
      "step": 900
    },
    {
      "epoch": 1.3846153846153846,
      "eval_loss": 0.8228450417518616,
      "eval_runtime": 2.0513,
      "eval_samples_per_second": 55.575,
      "eval_steps_per_second": 2.925,
      "step": 900
    },
    {
      "epoch": 1.4,
      "grad_norm": 3.7861831188201904,
      "learning_rate": 1.9396926207859085e-05,
      "loss": 0.3973,
      "step": 910
    },
    {
      "epoch": 1.4,
      "eval_loss": 0.8188081979751587,
      "eval_runtime": 2.0417,
      "eval_samples_per_second": 55.836,
      "eval_steps_per_second": 2.939,
      "step": 910
    },
    {
      "epoch": 1.4153846153846155,
      "grad_norm": 2.78849458694458,
      "learning_rate": 1.9370410910642473e-05,
      "loss": 0.333,
      "step": 920
    },
    {
      "epoch": 1.4153846153846155,
      "eval_loss": 0.8391227722167969,
      "eval_runtime": 2.0418,
      "eval_samples_per_second": 55.834,
      "eval_steps_per_second": 2.939,
      "step": 920
    },
    {
      "epoch": 1.4307692307692308,
      "grad_norm": 2.419405460357666,
      "learning_rate": 1.934334410971489e-05,
      "loss": 0.3946,
      "step": 930
    },
    {
      "epoch": 1.4307692307692308,
      "eval_loss": 0.8328570127487183,
      "eval_runtime": 2.0284,
      "eval_samples_per_second": 56.201,
      "eval_steps_per_second": 2.958,
      "step": 930
    },
    {
      "epoch": 1.4461538461538461,
      "grad_norm": 3.808756113052368,
      "learning_rate": 1.9315727398116516e-05,
      "loss": 0.4482,
      "step": 940
    },
    {
      "epoch": 1.4461538461538461,
      "eval_loss": 0.8310372829437256,
      "eval_runtime": 2.0424,
      "eval_samples_per_second": 55.817,
      "eval_steps_per_second": 2.938,
      "step": 940
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 4.97286319732666,
      "learning_rate": 1.9287562401253023e-05,
      "loss": 0.3848,
      "step": 950
    },
    {
      "epoch": 1.4615384615384617,
      "eval_loss": 0.8306027054786682,
      "eval_runtime": 2.0358,
      "eval_samples_per_second": 55.999,
      "eval_steps_per_second": 2.947,
      "step": 950
    },
    {
      "epoch": 1.476923076923077,
      "grad_norm": 3.4245073795318604,
      "learning_rate": 1.9258850776799875e-05,
      "loss": 0.391,
      "step": 960
    },
    {
      "epoch": 1.476923076923077,
      "eval_loss": 0.8288912773132324,
      "eval_runtime": 2.0384,
      "eval_samples_per_second": 55.925,
      "eval_steps_per_second": 2.943,
      "step": 960
    },
    {
      "epoch": 1.4923076923076923,
      "grad_norm": 1.8853943347930908,
      "learning_rate": 1.9229594214604782e-05,
      "loss": 0.3508,
      "step": 970
    },
    {
      "epoch": 1.4923076923076923,
      "eval_loss": 0.8303014039993286,
      "eval_runtime": 2.0396,
      "eval_samples_per_second": 55.894,
      "eval_steps_per_second": 2.942,
      "step": 970
    },
    {
      "epoch": 1.5076923076923077,
      "grad_norm": 2.889016628265381,
      "learning_rate": 1.9199794436588244e-05,
      "loss": 0.3516,
      "step": 980
    },
    {
      "epoch": 1.5076923076923077,
      "eval_loss": 0.8296997547149658,
      "eval_runtime": 2.0421,
      "eval_samples_per_second": 55.825,
      "eval_steps_per_second": 2.938,
      "step": 980
    },
    {
      "epoch": 1.523076923076923,
      "grad_norm": 1.8139469623565674,
      "learning_rate": 1.9169453196642197e-05,
      "loss": 0.3859,
      "step": 990
    },
    {
      "epoch": 1.523076923076923,
      "eval_loss": 0.8299453258514404,
      "eval_runtime": 2.0398,
      "eval_samples_per_second": 55.887,
      "eval_steps_per_second": 2.941,
      "step": 990
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 2.426495313644409,
      "learning_rate": 1.9138572280526795e-05,
      "loss": 0.4101,
      "step": 1000
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 0.8216718435287476,
      "eval_runtime": 2.0532,
      "eval_samples_per_second": 55.522,
      "eval_steps_per_second": 2.922,
      "step": 1000
    }
  ],
  "logging_steps": 10,
  "max_steps": 4550,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.808271792911155e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}