{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6459054209919262,
  "global_step": 1400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5.725190839694656e-06,
      "loss": 12.7187,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.1450381679389312e-05,
      "loss": 13.2124,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.6603053435114505e-05,
      "loss": 14.1489,
      "step": 30
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.1755725190839692e-05,
      "loss": 25.9008,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.6908396946564882e-05,
      "loss": 31.3861,
      "step": 50
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.263358778625954e-05,
      "loss": 8.9557,
      "step": 60
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.835877862595419e-05,
      "loss": 5.8443,
      "step": 70
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.408396946564885e-05,
      "loss": 4.3173,
      "step": 80
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.980916030534351e-05,
      "loss": 4.7444,
      "step": 90
    },
    {
      "epoch": 0.05,
      "learning_rate": 5.49618320610687e-05,
      "loss": 4.542,
      "step": 100
    },
    {
      "epoch": 0.05,
      "eval_loss": 3.579599618911743,
      "eval_runtime": 132.6751,
      "eval_samples_per_second": 34.822,
      "eval_steps_per_second": 1.093,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 0.05,
      "learning_rate": 6.0687022900763354e-05,
      "loss": 3.3873,
      "step": 110
    },
    {
      "epoch": 0.06,
      "learning_rate": 6.641221374045802e-05,
      "loss": 3.1783,
      "step": 120
    },
    {
      "epoch": 0.06,
      "learning_rate": 7.213740458015267e-05,
      "loss": 3.1672,
      "step": 130
    },
    {
      "epoch": 0.06,
      "learning_rate": 7.494113029827315e-05,
      "loss": 3.4915,
      "step": 140
    },
    {
      "epoch": 0.07,
      "learning_rate": 7.483516483516483e-05,
      "loss": 3.7276,
      "step": 150
    },
    {
      "epoch": 0.07,
      "learning_rate": 7.471742543171114e-05,
      "loss": 3.2435,
      "step": 160
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.459968602825744e-05,
      "loss": 3.0634,
      "step": 170
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.448194662480377e-05,
      "loss": 3.0859,
      "step": 180
    },
    {
      "epoch": 0.09,
      "learning_rate": 7.436420722135007e-05,
      "loss": 3.3592,
      "step": 190
    },
    {
      "epoch": 0.09,
      "learning_rate": 7.425824175824176e-05,
      "loss": 3.6018,
      "step": 200
    },
    {
      "epoch": 0.09,
      "eval_loss": 3.3504831790924072,
      "eval_runtime": 130.1255,
      "eval_samples_per_second": 35.504,
      "eval_steps_per_second": 1.114,
      "eval_wer": 1.0,
      "step": 200
    },
    {
      "epoch": 0.1,
      "learning_rate": 7.414050235478806e-05,
      "loss": 3.1612,
      "step": 210
    },
    {
      "epoch": 0.1,
      "learning_rate": 7.402276295133437e-05,
      "loss": 3.0375,
      "step": 220
    },
    {
      "epoch": 0.11,
      "learning_rate": 7.390502354788069e-05,
      "loss": 3.08,
      "step": 230
    },
    {
      "epoch": 0.11,
      "learning_rate": 7.3787284144427e-05,
      "loss": 3.2809,
      "step": 240
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.368131868131867e-05,
      "loss": 3.5351,
      "step": 250
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.356357927786499e-05,
      "loss": 3.01,
      "step": 260
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.34458398744113e-05,
      "loss": 3.0006,
      "step": 270
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.33281004709576e-05,
      "loss": 3.0397,
      "step": 280
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.321036106750392e-05,
      "loss": 3.1733,
      "step": 290
    },
    {
      "epoch": 0.14,
      "learning_rate": 7.310439560439559e-05,
      "loss": 3.5007,
      "step": 300
    },
    {
      "epoch": 0.14,
      "eval_loss": 3.068483591079712,
      "eval_runtime": 129.9806,
      "eval_samples_per_second": 35.544,
      "eval_steps_per_second": 1.116,
      "eval_wer": 1.0,
      "step": 300
    },
    {
      "epoch": 0.14,
      "learning_rate": 7.29866562009419e-05,
      "loss": 2.9984,
      "step": 310
    },
    {
      "epoch": 0.15,
      "learning_rate": 7.286891679748822e-05,
      "loss": 3.0006,
      "step": 320
    },
    {
      "epoch": 0.15,
      "learning_rate": 7.275117739403453e-05,
      "loss": 3.029,
      "step": 330
    },
    {
      "epoch": 0.16,
      "learning_rate": 7.263343799058085e-05,
      "loss": 3.158,
      "step": 340
    },
    {
      "epoch": 0.16,
      "learning_rate": 7.252747252747252e-05,
      "loss": 3.4909,
      "step": 350
    },
    {
      "epoch": 0.17,
      "learning_rate": 7.240973312401883e-05,
      "loss": 3.001,
      "step": 360
    },
    {
      "epoch": 0.17,
      "learning_rate": 7.229199372056515e-05,
      "loss": 2.9925,
      "step": 370
    },
    {
      "epoch": 0.18,
      "learning_rate": 7.217425431711145e-05,
      "loss": 3.0009,
      "step": 380
    },
    {
      "epoch": 0.18,
      "learning_rate": 7.205651491365776e-05,
      "loss": 3.1263,
      "step": 390
    },
    {
      "epoch": 0.18,
      "learning_rate": 7.195054945054944e-05,
      "loss": 3.421,
      "step": 400
    },
    {
      "epoch": 0.18,
      "eval_loss": 3.0357792377471924,
      "eval_runtime": 130.0328,
      "eval_samples_per_second": 35.529,
      "eval_steps_per_second": 1.115,
      "eval_wer": 1.0,
      "step": 400
    },
    {
      "epoch": 0.19,
      "learning_rate": 7.183281004709575e-05,
      "loss": 3.0018,
      "step": 410
    },
    {
      "epoch": 0.19,
      "learning_rate": 7.171507064364207e-05,
      "loss": 2.9947,
      "step": 420
    },
    {
      "epoch": 0.2,
      "learning_rate": 7.159733124018838e-05,
      "loss": 2.9935,
      "step": 430
    },
    {
      "epoch": 0.2,
      "learning_rate": 7.147959183673469e-05,
      "loss": 3.0818,
      "step": 440
    },
    {
      "epoch": 0.21,
      "learning_rate": 7.137362637362637e-05,
      "loss": 3.3386,
      "step": 450
    },
    {
      "epoch": 0.21,
      "learning_rate": 7.125588697017268e-05,
      "loss": 3.0803,
      "step": 460
    },
    {
      "epoch": 0.22,
      "learning_rate": 7.1138147566719e-05,
      "loss": 3.1655,
      "step": 470
    },
    {
      "epoch": 0.22,
      "learning_rate": 7.10204081632653e-05,
      "loss": 3.0899,
      "step": 480
    },
    {
      "epoch": 0.23,
      "learning_rate": 7.090266875981161e-05,
      "loss": 3.1638,
      "step": 490
    },
    {
      "epoch": 0.23,
      "learning_rate": 7.079670329670328e-05,
      "loss": 3.3518,
      "step": 500
    },
    {
      "epoch": 0.23,
      "eval_loss": 2.9412434101104736,
      "eval_runtime": 130.1575,
      "eval_samples_per_second": 35.495,
      "eval_steps_per_second": 1.114,
      "eval_wer": 1.0,
      "step": 500
    },
    {
      "epoch": 0.24,
      "learning_rate": 7.06789638932496e-05,
      "loss": 2.9352,
      "step": 510
    },
    {
      "epoch": 0.24,
      "learning_rate": 7.056122448979591e-05,
      "loss": 2.9405,
      "step": 520
    },
    {
      "epoch": 0.24,
      "learning_rate": 7.044348508634222e-05,
      "loss": 2.9125,
      "step": 530
    },
    {
      "epoch": 0.25,
      "learning_rate": 7.032574568288854e-05,
      "loss": 2.9401,
      "step": 540
    },
    {
      "epoch": 0.25,
      "learning_rate": 7.021978021978021e-05,
      "loss": 3.1443,
      "step": 550
    },
    {
      "epoch": 0.26,
      "learning_rate": 7.010204081632653e-05,
      "loss": 2.7692,
      "step": 560
    },
    {
      "epoch": 0.26,
      "learning_rate": 6.998430141287284e-05,
      "loss": 2.7574,
      "step": 570
    },
    {
      "epoch": 0.27,
      "learning_rate": 6.986656200941914e-05,
      "loss": 2.706,
      "step": 580
    },
    {
      "epoch": 0.27,
      "learning_rate": 6.974882260596545e-05,
      "loss": 2.7155,
      "step": 590
    },
    {
      "epoch": 0.28,
      "learning_rate": 6.964285714285713e-05,
      "loss": 2.8747,
      "step": 600
    },
    {
      "epoch": 0.28,
      "eval_loss": 2.5006003379821777,
      "eval_runtime": 129.3111,
      "eval_samples_per_second": 35.728,
      "eval_steps_per_second": 1.121,
      "eval_wer": 1.0,
      "step": 600
    },
    {
      "epoch": 0.28,
      "learning_rate": 6.952511773940345e-05,
      "loss": 2.5368,
      "step": 610
    },
    {
      "epoch": 0.29,
      "learning_rate": 6.940737833594976e-05,
      "loss": 2.5942,
      "step": 620
    },
    {
      "epoch": 0.29,
      "learning_rate": 6.928963893249607e-05,
      "loss": 2.4907,
      "step": 630
    },
    {
      "epoch": 0.3,
      "learning_rate": 6.917189952904238e-05,
      "loss": 2.472,
      "step": 640
    },
    {
      "epoch": 0.3,
      "learning_rate": 6.907770800627943e-05,
      "loss": 2.6826,
      "step": 650
    },
    {
      "epoch": 0.3,
      "learning_rate": 6.895996860282574e-05,
      "loss": 2.3167,
      "step": 660
    },
    {
      "epoch": 0.31,
      "learning_rate": 6.884222919937205e-05,
      "loss": 2.3457,
      "step": 670
    },
    {
      "epoch": 0.31,
      "learning_rate": 6.872448979591836e-05,
      "loss": 2.3532,
      "step": 680
    },
    {
      "epoch": 0.32,
      "learning_rate": 6.860675039246468e-05,
      "loss": 2.4822,
      "step": 690
    },
    {
      "epoch": 0.32,
      "learning_rate": 6.850078492935636e-05,
      "loss": 3.1002,
      "step": 700
    },
    {
      "epoch": 0.32,
      "eval_loss": 2.819200277328491,
      "eval_runtime": 129.5044,
      "eval_samples_per_second": 35.674,
      "eval_steps_per_second": 1.12,
      "eval_wer": 0.9999694058618369,
      "step": 700
    },
    {
      "epoch": 0.33,
      "learning_rate": 6.838304552590267e-05,
      "loss": 2.9139,
      "step": 710
    },
    {
      "epoch": 0.33,
      "learning_rate": 6.826530612244897e-05,
      "loss": 2.9957,
      "step": 720
    },
    {
      "epoch": 0.34,
      "learning_rate": 6.814756671899528e-05,
      "loss": 2.9842,
      "step": 730
    },
    {
      "epoch": 0.34,
      "learning_rate": 6.802982731554159e-05,
      "loss": 3.2804,
      "step": 740
    },
    {
      "epoch": 0.35,
      "learning_rate": 6.792386185243327e-05,
      "loss": 3.4944,
      "step": 750
    },
    {
      "epoch": 0.35,
      "learning_rate": 6.780612244897958e-05,
      "loss": 2.9905,
      "step": 760
    },
    {
      "epoch": 0.36,
      "learning_rate": 6.768838304552589e-05,
      "loss": 2.9692,
      "step": 770
    },
    {
      "epoch": 0.36,
      "learning_rate": 6.757064364207221e-05,
      "loss": 2.9747,
      "step": 780
    },
    {
      "epoch": 0.36,
      "learning_rate": 6.745290423861852e-05,
      "loss": 3.294,
      "step": 790
    },
    {
      "epoch": 0.37,
      "learning_rate": 6.73469387755102e-05,
      "loss": 3.5998,
      "step": 800
    },
    {
      "epoch": 0.37,
      "eval_loss": 3.059178590774536,
      "eval_runtime": 129.2337,
      "eval_samples_per_second": 35.749,
      "eval_steps_per_second": 1.122,
      "eval_wer": 1.0,
      "step": 800
    },
    {
      "epoch": 0.37,
      "learning_rate": 6.72291993720565e-05,
      "loss": 2.9902,
      "step": 810
    },
    {
      "epoch": 0.38,
      "learning_rate": 6.711145996860281e-05,
      "loss": 2.9644,
      "step": 820
    },
    {
      "epoch": 0.38,
      "learning_rate": 6.699372056514913e-05,
      "loss": 2.9529,
      "step": 830
    },
    {
      "epoch": 0.39,
      "learning_rate": 6.687598116169544e-05,
      "loss": 3.2737,
      "step": 840
    },
    {
      "epoch": 0.39,
      "learning_rate": 6.677001569858712e-05,
      "loss": 3.6534,
      "step": 850
    },
    {
      "epoch": 0.4,
      "learning_rate": 6.665227629513343e-05,
      "loss": 2.9664,
      "step": 860
    },
    {
      "epoch": 0.4,
      "learning_rate": 6.653453689167974e-05,
      "loss": 2.9463,
      "step": 870
    },
    {
      "epoch": 0.41,
      "learning_rate": 6.641679748822606e-05,
      "loss": 2.9591,
      "step": 880
    },
    {
      "epoch": 0.41,
      "learning_rate": 6.629905808477237e-05,
      "loss": 3.2701,
      "step": 890
    },
    {
      "epoch": 0.42,
      "learning_rate": 6.618131868131867e-05,
      "loss": 3.3784,
      "step": 900
    },
    {
      "epoch": 0.42,
      "eval_loss": 3.0334482192993164,
      "eval_runtime": 129.072,
      "eval_samples_per_second": 35.794,
      "eval_steps_per_second": 1.123,
      "eval_wer": 1.0,
      "step": 900
    },
    {
      "epoch": 0.42,
      "learning_rate": 6.606357927786498e-05,
      "loss": 2.9729,
      "step": 910
    },
    {
      "epoch": 0.42,
      "learning_rate": 6.594583987441129e-05,
      "loss": 2.9444,
      "step": 920
    },
    {
      "epoch": 0.43,
      "learning_rate": 6.582810047095761e-05,
      "loss": 2.9441,
      "step": 930
    },
    {
      "epoch": 0.43,
      "learning_rate": 6.571036106750392e-05,
      "loss": 3.2629,
      "step": 940
    },
    {
      "epoch": 0.44,
      "learning_rate": 6.56043956043956e-05,
      "loss": 3.6467,
      "step": 950
    },
    {
      "epoch": 0.44,
      "learning_rate": 6.548665620094191e-05,
      "loss": 2.9472,
      "step": 960
    },
    {
      "epoch": 0.45,
      "learning_rate": 6.536891679748822e-05,
      "loss": 2.9301,
      "step": 970
    },
    {
      "epoch": 0.45,
      "learning_rate": 6.525117739403452e-05,
      "loss": 2.9555,
      "step": 980
    },
    {
      "epoch": 0.46,
      "learning_rate": 6.513343799058084e-05,
      "loss": 3.2888,
      "step": 990
    },
    {
      "epoch": 0.46,
      "learning_rate": 6.502747252747253e-05,
      "loss": 3.6354,
      "step": 1000
    },
    {
      "epoch": 0.46,
      "eval_loss": 2.9995181560516357,
      "eval_runtime": 128.998,
      "eval_samples_per_second": 35.815,
      "eval_steps_per_second": 1.124,
      "eval_wer": 1.0,
      "step": 1000
    },
    {
      "epoch": 0.47,
      "learning_rate": 6.490973312401883e-05,
      "loss": 2.9439,
      "step": 1010
    },
    {
      "epoch": 0.47,
      "learning_rate": 6.479199372056514e-05,
      "loss": 2.9291,
      "step": 1020
    },
    {
      "epoch": 0.48,
      "learning_rate": 6.467425431711145e-05,
      "loss": 2.9709,
      "step": 1030
    },
    {
      "epoch": 0.48,
      "learning_rate": 6.455651491365777e-05,
      "loss": 3.3363,
      "step": 1040
    },
    {
      "epoch": 0.48,
      "learning_rate": 6.443877551020408e-05,
      "loss": 3.4812,
      "step": 1050
    },
    {
      "epoch": 0.49,
      "learning_rate": 6.432103610675038e-05,
      "loss": 2.9394,
      "step": 1060
    },
    {
      "epoch": 0.49,
      "learning_rate": 6.420329670329669e-05,
      "loss": 2.9355,
      "step": 1070
    },
    {
      "epoch": 0.5,
      "learning_rate": 6.4085557299843e-05,
      "loss": 2.9679,
      "step": 1080
    },
    {
      "epoch": 0.5,
      "learning_rate": 6.396781789638932e-05,
      "loss": 3.3439,
      "step": 1090
    },
    {
      "epoch": 0.51,
      "learning_rate": 6.3861852433281e-05,
      "loss": 3.7295,
      "step": 1100
    },
    {
      "epoch": 0.51,
      "eval_loss": 2.965841054916382,
      "eval_runtime": 129.5437,
      "eval_samples_per_second": 35.664,
      "eval_steps_per_second": 1.119,
      "eval_wer": 1.0,
      "step": 1100
    },
    {
      "epoch": 0.51,
      "learning_rate": 6.374411302982731e-05,
      "loss": 2.9341,
      "step": 1110
    },
    {
      "epoch": 0.52,
      "learning_rate": 6.362637362637362e-05,
      "loss": 2.923,
      "step": 1120
    },
    {
      "epoch": 0.52,
      "learning_rate": 6.350863422291993e-05,
      "loss": 2.9708,
      "step": 1130
    },
    {
      "epoch": 0.53,
      "learning_rate": 6.339089481946623e-05,
      "loss": 3.3851,
      "step": 1140
    },
    {
      "epoch": 0.53,
      "learning_rate": 6.328492935635793e-05,
      "loss": 3.7979,
      "step": 1150
    },
    {
      "epoch": 0.54,
      "learning_rate": 6.316718995290424e-05,
      "loss": 2.9171,
      "step": 1160
    },
    {
      "epoch": 0.54,
      "learning_rate": 6.304945054945054e-05,
      "loss": 2.9222,
      "step": 1170
    },
    {
      "epoch": 0.54,
      "learning_rate": 6.293171114599685e-05,
      "loss": 2.9591,
      "step": 1180
    },
    {
      "epoch": 0.55,
      "learning_rate": 6.281397174254316e-05,
      "loss": 3.3446,
      "step": 1190
    },
    {
      "epoch": 0.55,
      "learning_rate": 6.269623233908948e-05,
      "loss": 3.5604,
      "step": 1200
    },
    {
      "epoch": 0.55,
      "eval_loss": 2.9375498294830322,
      "eval_runtime": 129.4086,
      "eval_samples_per_second": 35.701,
      "eval_steps_per_second": 1.12,
      "eval_wer": 1.0,
      "step": 1200
    },
    {
      "epoch": 0.56,
      "learning_rate": 6.257849293563579e-05,
      "loss": 2.9104,
      "step": 1210
    },
    {
      "epoch": 0.56,
      "learning_rate": 6.24607535321821e-05,
      "loss": 2.9168,
      "step": 1220
    },
    {
      "epoch": 0.57,
      "learning_rate": 6.23430141287284e-05,
      "loss": 2.952,
      "step": 1230
    },
    {
      "epoch": 0.57,
      "learning_rate": 6.222527472527472e-05,
      "loss": 3.3335,
      "step": 1240
    },
    {
      "epoch": 0.58,
      "learning_rate": 6.21193092621664e-05,
      "loss": 3.7665,
      "step": 1250
    },
    {
      "epoch": 0.58,
      "learning_rate": 6.200156985871271e-05,
      "loss": 2.9118,
      "step": 1260
    },
    {
      "epoch": 0.59,
      "learning_rate": 6.188383045525902e-05,
      "loss": 2.918,
      "step": 1270
    },
    {
      "epoch": 0.59,
      "learning_rate": 6.176609105180533e-05,
      "loss": 2.9364,
      "step": 1280
    },
    {
      "epoch": 0.6,
      "learning_rate": 6.164835164835164e-05,
      "loss": 3.219,
      "step": 1290
    },
    {
      "epoch": 0.6,
      "learning_rate": 6.154238618524333e-05,
      "loss": 3.5977,
      "step": 1300
    },
    {
      "epoch": 0.6,
      "eval_loss": 2.94680118560791,
      "eval_runtime": 130.4848,
      "eval_samples_per_second": 35.406,
      "eval_steps_per_second": 1.111,
      "eval_wer": 0.9999694058618369,
      "step": 1300
    },
    {
      "epoch": 0.6,
      "learning_rate": 6.142464678178964e-05,
      "loss": 2.9167,
      "step": 1310
    },
    {
      "epoch": 0.61,
      "learning_rate": 6.130690737833595e-05,
      "loss": 2.9305,
      "step": 1320
    },
    {
      "epoch": 0.61,
      "learning_rate": 6.118916797488225e-05,
      "loss": 2.9303,
      "step": 1330
    },
    {
      "epoch": 0.62,
      "learning_rate": 6.107142857142856e-05,
      "loss": 3.1689,
      "step": 1340
    },
    {
      "epoch": 0.62,
      "learning_rate": 6.0965463108320243e-05,
      "loss": 3.4788,
      "step": 1350
    },
    {
      "epoch": 0.63,
      "learning_rate": 6.084772370486656e-05,
      "loss": 2.9461,
      "step": 1360
    },
    {
      "epoch": 0.63,
      "learning_rate": 6.072998430141287e-05,
      "loss": 2.9556,
      "step": 1370
    },
    {
      "epoch": 0.64,
      "learning_rate": 6.061224489795918e-05,
      "loss": 2.965,
      "step": 1380
    },
    {
      "epoch": 0.64,
      "learning_rate": 6.0494505494505494e-05,
      "loss": 3.1035,
      "step": 1390
    },
    {
      "epoch": 0.65,
      "learning_rate": 6.038854003139717e-05,
      "loss": 3.3783,
      "step": 1400
    },
    {
      "epoch": 0.65,
      "eval_loss": 3.044978141784668,
      "eval_runtime": 129.9307,
      "eval_samples_per_second": 35.557,
      "eval_steps_per_second": 1.116,
      "eval_wer": 1.0,
      "step": 1400
    }
  ],
  "max_steps": 6501,
  "num_train_epochs": 3,
  "total_flos": 2.64683689964352e+19,
  "trial_name": null,
  "trial_params": null
}