{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.904761904761905,
  "eval_steps": 500,
  "global_step": 520,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01904761904761905,
      "grad_norm": 4.625,
      "learning_rate": 3.846153846153847e-06,
      "loss": 3.0068,
      "step": 1
    },
    {
      "epoch": 0.09523809523809523,
      "grad_norm": 7.28125,
      "learning_rate": 1.923076923076923e-05,
      "loss": 3.002,
      "step": 5
    },
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 2.9375,
      "learning_rate": 3.846153846153846e-05,
      "loss": 2.8237,
      "step": 10
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 2.1875,
      "learning_rate": 5.769230769230769e-05,
      "loss": 2.4529,
      "step": 15
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 3.0625,
      "learning_rate": 7.692307692307693e-05,
      "loss": 2.1685,
      "step": 20
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 2.171875,
      "learning_rate": 9.615384615384617e-05,
      "loss": 1.9091,
      "step": 25
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.73046875,
      "learning_rate": 0.00011538461538461538,
      "loss": 1.6688,
      "step": 30
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.4921875,
      "learning_rate": 0.00013461538461538464,
      "loss": 1.4855,
      "step": 35
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 0.59375,
      "learning_rate": 0.00015384615384615385,
      "loss": 1.3575,
      "step": 40
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 0.734375,
      "learning_rate": 0.0001730769230769231,
      "loss": 1.2763,
      "step": 45
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019230769230769233,
      "loss": 1.2288,
      "step": 50
    },
    {
      "epoch": 0.9904761904761905,
      "eval_loss": 2.5090928077697754,
      "eval_runtime": 0.6698,
      "eval_samples_per_second": 14.929,
      "eval_steps_per_second": 1.493,
      "step": 52
    },
    {
      "epoch": 1.0476190476190477,
      "grad_norm": 0.3203125,
      "learning_rate": 0.00019997972289848503,
      "loss": 1.1945,
      "step": 55
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 0.45703125,
      "learning_rate": 0.00019985583705641418,
      "loss": 1.1593,
      "step": 60
    },
    {
      "epoch": 1.2380952380952381,
      "grad_norm": 0.76953125,
      "learning_rate": 0.00019961946980917456,
      "loss": 1.1262,
      "step": 65
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.5703125,
      "learning_rate": 0.0001992708874098054,
      "loss": 1.1179,
      "step": 70
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.357421875,
      "learning_rate": 0.0001988104825147528,
      "loss": 1.1067,
      "step": 75
    },
    {
      "epoch": 1.5238095238095237,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00019823877374156647,
      "loss": 1.0937,
      "step": 80
    },
    {
      "epoch": 1.619047619047619,
      "grad_norm": 0.8046875,
      "learning_rate": 0.00019755640508470942,
      "loss": 1.0861,
      "step": 85
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 0.400390625,
      "learning_rate": 0.00019676414519013781,
      "loss": 1.0766,
      "step": 90
    },
    {
      "epoch": 1.8095238095238095,
      "grad_norm": 0.34765625,
      "learning_rate": 0.00019586288648946947,
      "loss": 1.0619,
      "step": 95
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 0.314453125,
      "learning_rate": 0.00019485364419471454,
      "loss": 1.0651,
      "step": 100
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.2578125,
      "learning_rate": 0.00019373755515470254,
      "loss": 1.0492,
      "step": 105
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.4524765014648438,
      "eval_runtime": 0.4793,
      "eval_samples_per_second": 20.866,
      "eval_steps_per_second": 2.087,
      "step": 105
    },
    {
      "epoch": 2.0952380952380953,
      "grad_norm": 0.466796875,
      "learning_rate": 0.00019251587657449236,
      "loss": 1.0305,
      "step": 110
    },
    {
      "epoch": 2.1904761904761907,
      "grad_norm": 0.333984375,
      "learning_rate": 0.00019118998459920902,
      "loss": 1.0189,
      "step": 115
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 0.3515625,
      "learning_rate": 0.0001897613727639014,
      "loss": 1.0175,
      "step": 120
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0001882316503111678,
      "loss": 1.015,
      "step": 125
    },
    {
      "epoch": 2.4761904761904763,
      "grad_norm": 0.2314453125,
      "learning_rate": 0.00018660254037844388,
      "loss": 1.0152,
      "step": 130
    },
    {
      "epoch": 2.571428571428571,
      "grad_norm": 0.3203125,
      "learning_rate": 0.00018487587805699526,
      "loss": 1.0047,
      "step": 135
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.28125,
      "learning_rate": 0.00018305360832480117,
      "loss": 1.0046,
      "step": 140
    },
    {
      "epoch": 2.761904761904762,
      "grad_norm": 0.267578125,
      "learning_rate": 0.00018113778385565733,
      "loss": 1.0092,
      "step": 145
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.306640625,
      "learning_rate": 0.0001791305627069662,
      "loss": 0.9981,
      "step": 150
    },
    {
      "epoch": 2.9523809523809526,
      "grad_norm": 0.328125,
      "learning_rate": 0.00017703420588881946,
      "loss": 1.0025,
      "step": 155
    },
    {
      "epoch": 2.9904761904761905,
      "eval_loss": 2.4599478244781494,
      "eval_runtime": 0.4943,
      "eval_samples_per_second": 20.231,
      "eval_steps_per_second": 2.023,
      "step": 157
    },
    {
      "epoch": 3.0476190476190474,
      "grad_norm": 0.310546875,
      "learning_rate": 0.00017485107481711012,
      "loss": 0.9874,
      "step": 160
    },
    {
      "epoch": 3.142857142857143,
      "grad_norm": 0.3046875,
      "learning_rate": 0.00017258362865354426,
      "loss": 0.9747,
      "step": 165
    },
    {
      "epoch": 3.238095238095238,
      "grad_norm": 0.2353515625,
      "learning_rate": 0.00017023442153554777,
      "loss": 0.9763,
      "step": 170
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.318359375,
      "learning_rate": 0.0001678060996991891,
      "loss": 0.971,
      "step": 175
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0001653013984983585,
      "loss": 0.9666,
      "step": 180
    },
    {
      "epoch": 3.5238095238095237,
      "grad_norm": 0.703125,
      "learning_rate": 0.00016272313932356162,
      "loss": 0.9646,
      "step": 185
    },
    {
      "epoch": 3.619047619047619,
      "grad_norm": 0.796875,
      "learning_rate": 0.0001600742264237979,
      "loss": 0.9712,
      "step": 190
    },
    {
      "epoch": 3.7142857142857144,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0001573576436351046,
      "loss": 0.969,
      "step": 195
    },
    {
      "epoch": 3.8095238095238093,
      "grad_norm": 0.33984375,
      "learning_rate": 0.00015457645101945046,
      "loss": 0.9594,
      "step": 200
    },
    {
      "epoch": 3.9047619047619047,
      "grad_norm": 0.28515625,
      "learning_rate": 0.00015173378141776568,
      "loss": 0.9633,
      "step": 205
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.28515625,
      "learning_rate": 0.00014883283692099112,
      "loss": 0.9597,
      "step": 210
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.471402168273926,
      "eval_runtime": 0.4759,
      "eval_samples_per_second": 21.012,
      "eval_steps_per_second": 2.101,
      "step": 210
    },
    {
      "epoch": 4.095238095238095,
      "grad_norm": 0.31640625,
      "learning_rate": 0.00014587688526312143,
      "loss": 0.9325,
      "step": 215
    },
    {
      "epoch": 4.190476190476191,
      "grad_norm": 0.287109375,
      "learning_rate": 0.00014286925614030542,
      "loss": 0.933,
      "step": 220
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 0.2421875,
      "learning_rate": 0.0001398133374601501,
      "loss": 0.9404,
      "step": 225
    },
    {
      "epoch": 4.380952380952381,
      "grad_norm": 0.25390625,
      "learning_rate": 0.00013671257152545277,
      "loss": 0.9368,
      "step": 230
    },
    {
      "epoch": 4.476190476190476,
      "grad_norm": 0.2314453125,
      "learning_rate": 0.0001335704511566605,
      "loss": 0.934,
      "step": 235
    },
    {
      "epoch": 4.571428571428571,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0001303905157574247,
      "loss": 0.9391,
      "step": 240
    },
    {
      "epoch": 4.666666666666667,
      "grad_norm": 0.263671875,
      "learning_rate": 0.00012717634732768243,
      "loss": 0.9411,
      "step": 245
    },
    {
      "epoch": 4.761904761904762,
      "grad_norm": 0.263671875,
      "learning_rate": 0.0001239315664287558,
      "loss": 0.9428,
      "step": 250
    },
    {
      "epoch": 4.857142857142857,
      "grad_norm": 0.291015625,
      "learning_rate": 0.00012065982810501404,
      "loss": 0.9332,
      "step": 255
    },
    {
      "epoch": 4.9523809523809526,
      "grad_norm": 0.24609375,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.9394,
      "step": 260
    },
    {
      "epoch": 4.9904761904761905,
      "eval_loss": 2.486337423324585,
      "eval_runtime": 0.6707,
      "eval_samples_per_second": 14.909,
      "eval_steps_per_second": 1.491,
      "step": 262
    },
    {
      "epoch": 5.0476190476190474,
      "grad_norm": 0.48828125,
      "learning_rate": 0.00011405024703850929,
      "loss": 0.9283,
      "step": 265
    },
    {
      "epoch": 5.142857142857143,
      "grad_norm": 0.37890625,
      "learning_rate": 0.00011071984957874479,
      "loss": 0.9202,
      "step": 270
    },
    {
      "epoch": 5.238095238095238,
      "grad_norm": 0.25,
      "learning_rate": 0.00010737737687351284,
      "loss": 0.916,
      "step": 275
    },
    {
      "epoch": 5.333333333333333,
      "grad_norm": 0.25390625,
      "learning_rate": 0.00010402659401094152,
      "loss": 0.9109,
      "step": 280
    },
    {
      "epoch": 5.428571428571429,
      "grad_norm": 0.2734375,
      "learning_rate": 0.00010067127544003563,
      "loss": 0.9156,
      "step": 285
    },
    {
      "epoch": 5.523809523809524,
      "grad_norm": 0.259765625,
      "learning_rate": 9.73152007189939e-05,
      "loss": 0.9139,
      "step": 290
    },
    {
      "epoch": 5.619047619047619,
      "grad_norm": 0.234375,
      "learning_rate": 9.396215025777139e-05,
      "loss": 0.9166,
      "step": 295
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 0.2314453125,
      "learning_rate": 9.061590105968208e-05,
      "loss": 0.9162,
      "step": 300
    },
    {
      "epoch": 5.809523809523809,
      "grad_norm": 0.328125,
      "learning_rate": 8.728022246683894e-05,
      "loss": 0.9079,
      "step": 305
    },
    {
      "epoch": 5.904761904761905,
      "grad_norm": 0.263671875,
      "learning_rate": 8.395887191422397e-05,
      "loss": 0.9135,
      "step": 310
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.2578125,
      "learning_rate": 8.065559069717088e-05,
      "loss": 0.912,
      "step": 315
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.5085232257843018,
      "eval_runtime": 0.4762,
      "eval_samples_per_second": 21.0,
      "eval_steps_per_second": 2.1,
      "step": 315
    },
    {
      "epoch": 6.095238095238095,
      "grad_norm": 0.2236328125,
      "learning_rate": 7.73740997570278e-05,
      "loss": 0.8989,
      "step": 320
    },
    {
      "epoch": 6.190476190476191,
      "grad_norm": 0.2578125,
      "learning_rate": 7.411809548974792e-05,
      "loss": 0.897,
      "step": 325
    },
    {
      "epoch": 6.285714285714286,
      "grad_norm": 0.31640625,
      "learning_rate": 7.089124558212871e-05,
      "loss": 0.9014,
      "step": 330
    },
    {
      "epoch": 6.380952380952381,
      "grad_norm": 0.353515625,
      "learning_rate": 6.769718488039023e-05,
      "loss": 0.8964,
      "step": 335
    },
    {
      "epoch": 6.476190476190476,
      "grad_norm": 0.267578125,
      "learning_rate": 6.453951129574644e-05,
      "loss": 0.894,
      "step": 340
    },
    {
      "epoch": 6.571428571428571,
      "grad_norm": 0.25390625,
      "learning_rate": 6.142178175158149e-05,
      "loss": 0.9002,
      "step": 345
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.2294921875,
      "learning_rate": 5.834750817679606e-05,
      "loss": 0.8989,
      "step": 350
    },
    {
      "epoch": 6.761904761904762,
      "grad_norm": 0.2265625,
      "learning_rate": 5.5320153549837415e-05,
      "loss": 0.8996,
      "step": 355
    },
    {
      "epoch": 6.857142857142857,
      "grad_norm": 0.23046875,
      "learning_rate": 5.234312799786921e-05,
      "loss": 0.9,
      "step": 360
    },
    {
      "epoch": 6.9523809523809526,
      "grad_norm": 0.2412109375,
      "learning_rate": 4.9419784955474524e-05,
      "loss": 0.8901,
      "step": 365
    },
    {
      "epoch": 6.9904761904761905,
      "eval_loss": 2.517164707183838,
      "eval_runtime": 0.6714,
      "eval_samples_per_second": 14.895,
      "eval_steps_per_second": 1.489,
      "step": 367
    },
    {
      "epoch": 7.0476190476190474,
      "grad_norm": 0.21484375,
      "learning_rate": 4.6553417387219886e-05,
      "loss": 0.8844,
      "step": 370
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 0.2216796875,
      "learning_rate": 4.374725407833532e-05,
      "loss": 0.878,
      "step": 375
    },
    {
      "epoch": 7.238095238095238,
      "grad_norm": 0.2294921875,
      "learning_rate": 4.100445599768774e-05,
      "loss": 0.8811,
      "step": 380
    },
    {
      "epoch": 7.333333333333333,
      "grad_norm": 0.212890625,
      "learning_rate": 3.832811273714569e-05,
      "loss": 0.8853,
      "step": 385
    },
    {
      "epoch": 7.428571428571429,
      "grad_norm": 0.34375,
      "learning_rate": 3.5721239031346066e-05,
      "loss": 0.8852,
      "step": 390
    },
    {
      "epoch": 7.523809523809524,
      "grad_norm": 0.224609375,
      "learning_rate": 3.318677136178228e-05,
      "loss": 0.8829,
      "step": 395
    },
    {
      "epoch": 7.619047619047619,
      "grad_norm": 0.232421875,
      "learning_rate": 3.072756464904006e-05,
      "loss": 0.8899,
      "step": 400
    },
    {
      "epoch": 7.714285714285714,
      "grad_norm": 0.302734375,
      "learning_rate": 2.8346389036906828e-05,
      "loss": 0.888,
      "step": 405
    },
    {
      "epoch": 7.809523809523809,
      "grad_norm": 0.21875,
      "learning_rate": 2.6045926771976303e-05,
      "loss": 0.8876,
      "step": 410
    },
    {
      "epoch": 7.904761904761905,
      "grad_norm": 0.21875,
      "learning_rate": 2.382876918226409e-05,
      "loss": 0.8924,
      "step": 415
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.2138671875,
      "learning_rate": 2.1697413758237784e-05,
      "loss": 0.8887,
      "step": 420
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.529445171356201,
      "eval_runtime": 0.4793,
      "eval_samples_per_second": 20.866,
      "eval_steps_per_second": 2.087,
      "step": 420
    },
    {
      "epoch": 8.095238095238095,
      "grad_norm": 0.26171875,
      "learning_rate": 1.965426133954854e-05,
      "loss": 0.8761,
      "step": 425
    },
    {
      "epoch": 8.19047619047619,
      "grad_norm": 0.2177734375,
      "learning_rate": 1.7701613410634365e-05,
      "loss": 0.8847,
      "step": 430
    },
    {
      "epoch": 8.285714285714286,
      "grad_norm": 0.2197265625,
      "learning_rate": 1.584166950824061e-05,
      "loss": 0.8831,
      "step": 435
    },
    {
      "epoch": 8.380952380952381,
      "grad_norm": 0.2080078125,
      "learning_rate": 1.4076524743778319e-05,
      "loss": 0.8769,
      "step": 440
    },
    {
      "epoch": 8.476190476190476,
      "grad_norm": 0.205078125,
      "learning_rate": 1.2408167443311214e-05,
      "loss": 0.8773,
      "step": 445
    },
    {
      "epoch": 8.571428571428571,
      "grad_norm": 0.21875,
      "learning_rate": 1.083847690782972e-05,
      "loss": 0.8772,
      "step": 450
    },
    {
      "epoch": 8.666666666666666,
      "grad_norm": 0.236328125,
      "learning_rate": 9.369221296335006e-06,
      "loss": 0.8846,
      "step": 455
    },
    {
      "epoch": 8.761904761904763,
      "grad_norm": 0.216796875,
      "learning_rate": 8.002055634117578e-06,
      "loss": 0.8759,
      "step": 460
    },
    {
      "epoch": 8.857142857142858,
      "grad_norm": 0.203125,
      "learning_rate": 6.738519948473976e-06,
      "loss": 0.8798,
      "step": 465
    },
    {
      "epoch": 8.952380952380953,
      "grad_norm": 0.20703125,
      "learning_rate": 5.580037533961546e-06,
      "loss": 0.8816,
      "step": 470
    },
    {
      "epoch": 8.99047619047619,
      "eval_loss": 2.531066417694092,
      "eval_runtime": 0.4939,
      "eval_samples_per_second": 20.246,
      "eval_steps_per_second": 2.025,
      "step": 472
    },
    {
      "epoch": 9.047619047619047,
      "grad_norm": 0.203125,
      "learning_rate": 4.527913349145441e-06,
      "loss": 0.878,
      "step": 475
    },
    {
      "epoch": 9.142857142857142,
      "grad_norm": 0.203125,
      "learning_rate": 3.5833325466437694e-06,
      "loss": 0.8736,
      "step": 480
    },
    {
      "epoch": 9.238095238095237,
      "grad_norm": 0.208984375,
      "learning_rate": 2.7473591381266708e-06,
      "loss": 0.8763,
      "step": 485
    },
    {
      "epoch": 9.333333333333334,
      "grad_norm": 0.20703125,
      "learning_rate": 2.0209347957732328e-06,
      "loss": 0.8832,
      "step": 490
    },
    {
      "epoch": 9.428571428571429,
      "grad_norm": 0.203125,
      "learning_rate": 1.404877791536141e-06,
      "loss": 0.879,
      "step": 495
    },
    {
      "epoch": 9.523809523809524,
      "grad_norm": 0.19921875,
      "learning_rate": 8.998820754091531e-07,
      "loss": 0.8787,
      "step": 500
    },
    {
      "epoch": 9.619047619047619,
      "grad_norm": 0.201171875,
      "learning_rate": 5.065164937354428e-07,
      "loss": 0.8782,
      "step": 505
    },
    {
      "epoch": 9.714285714285714,
      "grad_norm": 0.2080078125,
      "learning_rate": 2.2522414843748618e-07,
      "loss": 0.8803,
      "step": 510
    },
    {
      "epoch": 9.80952380952381,
      "grad_norm": 0.201171875,
      "learning_rate": 5.632189789027687e-08,
      "loss": 0.8719,
      "step": 515
    },
    {
      "epoch": 9.904761904761905,
      "grad_norm": 0.19921875,
      "learning_rate": 0.0,
      "loss": 0.8797,
      "step": 520
    },
    {
      "epoch": 9.904761904761905,
      "eval_loss": 2.532215118408203,
      "eval_runtime": 0.4799,
      "eval_samples_per_second": 20.838,
      "eval_steps_per_second": 2.084,
      "step": 520
    },
    {
      "epoch": 9.904761904761905,
      "step": 520,
      "total_flos": 8.265211656850637e+17,
      "train_loss": 1.0387523100926326,
      "train_runtime": 3563.5293,
      "train_samples_per_second": 18.703,
      "train_steps_per_second": 0.146
    }
  ],
  "logging_steps": 5,
  "max_steps": 520,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.265211656850637e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}