{
  "best_metric": 0.8048780487804879,
  "best_model_checkpoint": "MAE-CT-CPC-Dicotomized-v4-early-stop/checkpoint-648",
  "epoch": 15.05625,
  "eval_steps": 500,
  "global_step": 1296,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 6.944444444444446e-07,
      "loss": 0.5724,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.3888888888888892e-06,
      "loss": 0.633,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.0833333333333334e-06,
      "loss": 0.5686,
      "step": 30
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.605,
      "step": 40
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.4722222222222224e-06,
      "loss": 0.7115,
      "step": 50
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.6218,
      "step": 60
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.861111111111111e-06,
      "loss": 0.6258,
      "step": 70
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.555555555555557e-06,
      "loss": 0.5342,
      "step": 80
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.6829268292682927,
      "eval_loss": 0.6496201753616333,
      "eval_runtime": 8.1028,
      "eval_samples_per_second": 5.06,
      "eval_steps_per_second": 1.358,
      "step": 81
    },
    {
      "epoch": 1.01,
      "learning_rate": 6.25e-06,
      "loss": 0.6638,
      "step": 90
    },
    {
      "epoch": 1.01,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.5117,
      "step": 100
    },
    {
      "epoch": 1.02,
      "learning_rate": 7.638888888888888e-06,
      "loss": 0.6552,
      "step": 110
    },
    {
      "epoch": 1.03,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.5231,
      "step": 120
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.027777777777779e-06,
      "loss": 0.6073,
      "step": 130
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.722222222222223e-06,
      "loss": 0.6471,
      "step": 140
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.953703703703704e-06,
      "loss": 0.6375,
      "step": 150
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.876543209876543e-06,
      "loss": 0.6266,
      "step": 160
    },
    {
      "epoch": 1.06,
      "eval_accuracy": 0.6829268292682927,
      "eval_loss": 0.5853927135467529,
      "eval_runtime": 8.0931,
      "eval_samples_per_second": 5.066,
      "eval_steps_per_second": 1.359,
      "step": 162
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.799382716049384e-06,
      "loss": 0.5042,
      "step": 170
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.722222222222223e-06,
      "loss": 0.4928,
      "step": 180
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.645061728395062e-06,
      "loss": 0.5231,
      "step": 190
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.567901234567902e-06,
      "loss": 0.5299,
      "step": 200
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.490740740740741e-06,
      "loss": 0.6126,
      "step": 210
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.413580246913581e-06,
      "loss": 0.4196,
      "step": 220
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.33641975308642e-06,
      "loss": 0.9175,
      "step": 230
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.25925925925926e-06,
      "loss": 0.6599,
      "step": 240
    },
    {
      "epoch": 2.06,
      "eval_accuracy": 0.6829268292682927,
      "eval_loss": 0.5205796957015991,
      "eval_runtime": 8.4167,
      "eval_samples_per_second": 4.871,
      "eval_steps_per_second": 1.307,
      "step": 243
    },
    {
      "epoch": 3.0,
      "learning_rate": 9.1820987654321e-06,
      "loss": 0.5163,
      "step": 250
    },
    {
      "epoch": 3.01,
      "learning_rate": 9.10493827160494e-06,
      "loss": 0.6428,
      "step": 260
    },
    {
      "epoch": 3.02,
      "learning_rate": 9.027777777777779e-06,
      "loss": 0.5963,
      "step": 270
    },
    {
      "epoch": 3.03,
      "learning_rate": 8.950617283950618e-06,
      "loss": 0.5579,
      "step": 280
    },
    {
      "epoch": 3.03,
      "learning_rate": 8.873456790123458e-06,
      "loss": 0.6436,
      "step": 290
    },
    {
      "epoch": 3.04,
      "learning_rate": 8.796296296296297e-06,
      "loss": 0.6368,
      "step": 300
    },
    {
      "epoch": 3.05,
      "learning_rate": 8.719135802469136e-06,
      "loss": 0.5194,
      "step": 310
    },
    {
      "epoch": 3.05,
      "learning_rate": 8.641975308641975e-06,
      "loss": 0.877,
      "step": 320
    },
    {
      "epoch": 3.06,
      "eval_accuracy": 0.6097560975609756,
      "eval_loss": 0.5995270013809204,
      "eval_runtime": 8.3891,
      "eval_samples_per_second": 4.887,
      "eval_steps_per_second": 1.311,
      "step": 324
    },
    {
      "epoch": 4.0,
      "learning_rate": 8.564814814814816e-06,
      "loss": 0.5644,
      "step": 330
    },
    {
      "epoch": 4.01,
      "learning_rate": 8.487654320987654e-06,
      "loss": 0.4776,
      "step": 340
    },
    {
      "epoch": 4.02,
      "learning_rate": 8.410493827160495e-06,
      "loss": 0.5117,
      "step": 350
    },
    {
      "epoch": 4.03,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.4388,
      "step": 360
    },
    {
      "epoch": 4.03,
      "learning_rate": 8.256172839506174e-06,
      "loss": 0.5065,
      "step": 370
    },
    {
      "epoch": 4.04,
      "learning_rate": 8.179012345679013e-06,
      "loss": 0.5147,
      "step": 380
    },
    {
      "epoch": 4.05,
      "learning_rate": 8.101851851851854e-06,
      "loss": 0.6924,
      "step": 390
    },
    {
      "epoch": 4.05,
      "learning_rate": 8.024691358024692e-06,
      "loss": 0.653,
      "step": 400
    },
    {
      "epoch": 4.06,
      "eval_accuracy": 0.7560975609756098,
      "eval_loss": 0.4907689392566681,
      "eval_runtime": 7.7719,
      "eval_samples_per_second": 5.275,
      "eval_steps_per_second": 1.415,
      "step": 405
    },
    {
      "epoch": 5.0,
      "learning_rate": 7.947530864197531e-06,
      "loss": 0.536,
      "step": 410
    },
    {
      "epoch": 5.01,
      "learning_rate": 7.870370370370372e-06,
      "loss": 0.4254,
      "step": 420
    },
    {
      "epoch": 5.02,
      "learning_rate": 7.79320987654321e-06,
      "loss": 0.4454,
      "step": 430
    },
    {
      "epoch": 5.02,
      "learning_rate": 7.71604938271605e-06,
      "loss": 0.4594,
      "step": 440
    },
    {
      "epoch": 5.03,
      "learning_rate": 7.638888888888888e-06,
      "loss": 0.4811,
      "step": 450
    },
    {
      "epoch": 5.04,
      "learning_rate": 7.561728395061729e-06,
      "loss": 0.4024,
      "step": 460
    },
    {
      "epoch": 5.05,
      "learning_rate": 7.484567901234569e-06,
      "loss": 0.3778,
      "step": 470
    },
    {
      "epoch": 5.05,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.7604,
      "step": 480
    },
    {
      "epoch": 5.06,
      "eval_accuracy": 0.7804878048780488,
      "eval_loss": 0.49357670545578003,
      "eval_runtime": 8.1369,
      "eval_samples_per_second": 5.039,
      "eval_steps_per_second": 1.352,
      "step": 486
    },
    {
      "epoch": 6.0,
      "learning_rate": 7.330246913580248e-06,
      "loss": 0.5151,
      "step": 490
    },
    {
      "epoch": 6.01,
      "learning_rate": 7.253086419753087e-06,
      "loss": 0.3392,
      "step": 500
    },
    {
      "epoch": 6.02,
      "learning_rate": 7.1759259259259266e-06,
      "loss": 0.5844,
      "step": 510
    },
    {
      "epoch": 6.02,
      "learning_rate": 7.098765432098766e-06,
      "loss": 0.5688,
      "step": 520
    },
    {
      "epoch": 6.03,
      "learning_rate": 7.021604938271606e-06,
      "loss": 0.247,
      "step": 530
    },
    {
      "epoch": 6.04,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.7517,
      "step": 540
    },
    {
      "epoch": 6.04,
      "learning_rate": 6.867283950617285e-06,
      "loss": 0.3608,
      "step": 550
    },
    {
      "epoch": 6.05,
      "learning_rate": 6.790123456790124e-06,
      "loss": 0.4795,
      "step": 560
    },
    {
      "epoch": 6.06,
      "eval_accuracy": 0.6829268292682927,
      "eval_loss": 0.9527755379676819,
      "eval_runtime": 8.3829,
      "eval_samples_per_second": 4.891,
      "eval_steps_per_second": 1.312,
      "step": 567
    },
    {
      "epoch": 7.0,
      "learning_rate": 6.712962962962963e-06,
      "loss": 0.3914,
      "step": 570
    },
    {
      "epoch": 7.01,
      "learning_rate": 6.635802469135803e-06,
      "loss": 0.4056,
      "step": 580
    },
    {
      "epoch": 7.02,
      "learning_rate": 6.558641975308642e-06,
      "loss": 0.4884,
      "step": 590
    },
    {
      "epoch": 7.02,
      "learning_rate": 6.481481481481482e-06,
      "loss": 0.3237,
      "step": 600
    },
    {
      "epoch": 7.03,
      "learning_rate": 6.404320987654321e-06,
      "loss": 0.4742,
      "step": 610
    },
    {
      "epoch": 7.04,
      "learning_rate": 6.3271604938271615e-06,
      "loss": 0.4839,
      "step": 620
    },
    {
      "epoch": 7.04,
      "learning_rate": 6.25e-06,
      "loss": 0.265,
      "step": 630
    },
    {
      "epoch": 7.05,
      "learning_rate": 6.17283950617284e-06,
      "loss": 0.278,
      "step": 640
    },
    {
      "epoch": 7.06,
      "eval_accuracy": 0.8048780487804879,
      "eval_loss": 0.5564919710159302,
      "eval_runtime": 8.4164,
      "eval_samples_per_second": 4.871,
      "eval_steps_per_second": 1.307,
      "step": 648
    },
    {
      "epoch": 8.0,
      "learning_rate": 6.09567901234568e-06,
      "loss": 0.1885,
      "step": 650
    },
    {
      "epoch": 8.01,
      "learning_rate": 6.018518518518519e-06,
      "loss": 0.2307,
      "step": 660
    },
    {
      "epoch": 8.02,
      "learning_rate": 5.941358024691358e-06,
      "loss": 0.4786,
      "step": 670
    },
    {
      "epoch": 8.02,
      "learning_rate": 5.864197530864199e-06,
      "loss": 0.3708,
      "step": 680
    },
    {
      "epoch": 8.03,
      "learning_rate": 5.787037037037038e-06,
      "loss": 0.5062,
      "step": 690
    },
    {
      "epoch": 8.04,
      "learning_rate": 5.7098765432098764e-06,
      "loss": 0.7825,
      "step": 700
    },
    {
      "epoch": 8.04,
      "learning_rate": 5.632716049382716e-06,
      "loss": 0.1902,
      "step": 710
    },
    {
      "epoch": 8.05,
      "learning_rate": 5.555555555555557e-06,
      "loss": 0.3548,
      "step": 720
    },
    {
      "epoch": 8.06,
      "eval_accuracy": 0.7560975609756098,
      "eval_loss": 0.5855191946029663,
      "eval_runtime": 7.7883,
      "eval_samples_per_second": 5.264,
      "eval_steps_per_second": 1.412,
      "step": 729
    },
    {
      "epoch": 9.0,
      "learning_rate": 5.4783950617283955e-06,
      "loss": 0.2654,
      "step": 730
    },
    {
      "epoch": 9.01,
      "learning_rate": 5.401234567901234e-06,
      "loss": 0.2532,
      "step": 740
    },
    {
      "epoch": 9.01,
      "learning_rate": 5.324074074074075e-06,
      "loss": 0.243,
      "step": 750
    },
    {
      "epoch": 9.02,
      "learning_rate": 5.246913580246914e-06,
      "loss": 0.4346,
      "step": 760
    },
    {
      "epoch": 9.03,
      "learning_rate": 5.1697530864197534e-06,
      "loss": 0.1327,
      "step": 770
    },
    {
      "epoch": 9.04,
      "learning_rate": 5.092592592592593e-06,
      "loss": 0.3484,
      "step": 780
    },
    {
      "epoch": 9.04,
      "learning_rate": 5.015432098765433e-06,
      "loss": 0.4213,
      "step": 790
    },
    {
      "epoch": 9.05,
      "learning_rate": 4.938271604938272e-06,
      "loss": 0.3673,
      "step": 800
    },
    {
      "epoch": 9.06,
      "learning_rate": 4.861111111111111e-06,
      "loss": 0.4386,
      "step": 810
    },
    {
      "epoch": 9.06,
      "eval_accuracy": 0.7560975609756098,
      "eval_loss": 0.6577650904655457,
      "eval_runtime": 8.1212,
      "eval_samples_per_second": 5.049,
      "eval_steps_per_second": 1.354,
      "step": 810
    },
    {
      "epoch": 10.01,
      "learning_rate": 4.783950617283951e-06,
      "loss": 0.5757,
      "step": 820
    },
    {
      "epoch": 10.01,
      "learning_rate": 4.706790123456791e-06,
      "loss": 0.2701,
      "step": 830
    },
    {
      "epoch": 10.02,
      "learning_rate": 4.62962962962963e-06,
      "loss": 0.1286,
      "step": 840
    },
    {
      "epoch": 10.03,
      "learning_rate": 4.55246913580247e-06,
      "loss": 0.2345,
      "step": 850
    },
    {
      "epoch": 10.03,
      "learning_rate": 4.475308641975309e-06,
      "loss": 0.1286,
      "step": 860
    },
    {
      "epoch": 10.04,
      "learning_rate": 4.398148148148149e-06,
      "loss": 0.3579,
      "step": 870
    },
    {
      "epoch": 10.05,
      "learning_rate": 4.3209876543209875e-06,
      "loss": 0.408,
      "step": 880
    },
    {
      "epoch": 10.06,
      "learning_rate": 4.243827160493827e-06,
      "loss": 0.3007,
      "step": 890
    },
    {
      "epoch": 10.06,
      "eval_accuracy": 0.7804878048780488,
      "eval_loss": 0.6622430682182312,
      "eval_runtime": 8.0725,
      "eval_samples_per_second": 5.079,
      "eval_steps_per_second": 1.363,
      "step": 891
    },
    {
      "epoch": 11.01,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.4887,
      "step": 900
    },
    {
      "epoch": 11.01,
      "learning_rate": 4.0895061728395066e-06,
      "loss": 0.1813,
      "step": 910
    },
    {
      "epoch": 11.02,
      "learning_rate": 4.012345679012346e-06,
      "loss": 0.1648,
      "step": 920
    },
    {
      "epoch": 11.03,
      "learning_rate": 3.935185185185186e-06,
      "loss": 0.4133,
      "step": 930
    },
    {
      "epoch": 11.03,
      "learning_rate": 3.858024691358025e-06,
      "loss": 0.1691,
      "step": 940
    },
    {
      "epoch": 11.04,
      "learning_rate": 3.7808641975308645e-06,
      "loss": 0.1626,
      "step": 950
    },
    {
      "epoch": 11.05,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.2215,
      "step": 960
    },
    {
      "epoch": 11.05,
      "learning_rate": 3.6265432098765434e-06,
      "loss": 0.313,
      "step": 970
    },
    {
      "epoch": 11.06,
      "eval_accuracy": 0.7560975609756098,
      "eval_loss": 0.8349580764770508,
      "eval_runtime": 8.0234,
      "eval_samples_per_second": 5.11,
      "eval_steps_per_second": 1.371,
      "step": 972
    },
    {
      "epoch": 12.01,
      "learning_rate": 3.549382716049383e-06,
      "loss": 0.4007,
      "step": 980
    },
    {
      "epoch": 12.01,
      "learning_rate": 3.4722222222222224e-06,
      "loss": 0.2461,
      "step": 990
    },
    {
      "epoch": 12.02,
      "learning_rate": 3.395061728395062e-06,
      "loss": 0.2541,
      "step": 1000
    },
    {
      "epoch": 12.03,
      "learning_rate": 3.3179012345679013e-06,
      "loss": 0.3941,
      "step": 1010
    },
    {
      "epoch": 12.03,
      "learning_rate": 3.240740740740741e-06,
      "loss": 0.0845,
      "step": 1020
    },
    {
      "epoch": 12.04,
      "learning_rate": 3.1635802469135807e-06,
      "loss": 0.2944,
      "step": 1030
    },
    {
      "epoch": 12.05,
      "learning_rate": 3.08641975308642e-06,
      "loss": 0.2391,
      "step": 1040
    },
    {
      "epoch": 12.05,
      "learning_rate": 3.0092592592592597e-06,
      "loss": 0.0554,
      "step": 1050
    },
    {
      "epoch": 12.06,
      "eval_accuracy": 0.7073170731707317,
      "eval_loss": 1.0043153762817383,
      "eval_runtime": 7.5796,
      "eval_samples_per_second": 5.409,
      "eval_steps_per_second": 1.451,
      "step": 1053
    },
    {
      "epoch": 13.0,
      "learning_rate": 2.9320987654320994e-06,
      "loss": 0.2887,
      "step": 1060
    },
    {
      "epoch": 13.01,
      "learning_rate": 2.8549382716049382e-06,
      "loss": 0.4346,
      "step": 1070
    },
    {
      "epoch": 13.02,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.3038,
      "step": 1080
    },
    {
      "epoch": 13.03,
      "learning_rate": 2.700617283950617e-06,
      "loss": 0.1073,
      "step": 1090
    },
    {
      "epoch": 13.03,
      "learning_rate": 2.623456790123457e-06,
      "loss": 0.2334,
      "step": 1100
    },
    {
      "epoch": 13.04,
      "learning_rate": 2.5462962962962966e-06,
      "loss": 0.34,
      "step": 1110
    },
    {
      "epoch": 13.05,
      "learning_rate": 2.469135802469136e-06,
      "loss": 0.0705,
      "step": 1120
    },
    {
      "epoch": 13.05,
      "learning_rate": 2.3919753086419755e-06,
      "loss": 0.2804,
      "step": 1130
    },
    {
      "epoch": 13.06,
      "eval_accuracy": 0.7073170731707317,
      "eval_loss": 1.0246809720993042,
      "eval_runtime": 7.4342,
      "eval_samples_per_second": 5.515,
      "eval_steps_per_second": 1.48,
      "step": 1134
    },
    {
      "epoch": 14.0,
      "learning_rate": 2.314814814814815e-06,
      "loss": 0.0367,
      "step": 1140
    },
    {
      "epoch": 14.01,
      "learning_rate": 2.2376543209876545e-06,
      "loss": 0.2872,
      "step": 1150
    },
    {
      "epoch": 14.02,
      "learning_rate": 2.1604938271604937e-06,
      "loss": 0.0806,
      "step": 1160
    },
    {
      "epoch": 14.03,
      "learning_rate": 2.0833333333333334e-06,
      "loss": 0.1376,
      "step": 1170
    },
    {
      "epoch": 14.03,
      "learning_rate": 2.006172839506173e-06,
      "loss": 0.466,
      "step": 1180
    },
    {
      "epoch": 14.04,
      "learning_rate": 1.9290123456790124e-06,
      "loss": 0.2337,
      "step": 1190
    },
    {
      "epoch": 14.05,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.3175,
      "step": 1200
    },
    {
      "epoch": 14.05,
      "learning_rate": 1.7746913580246916e-06,
      "loss": 0.1424,
      "step": 1210
    },
    {
      "epoch": 14.06,
      "eval_accuracy": 0.7804878048780488,
      "eval_loss": 0.8541743755340576,
      "eval_runtime": 7.7108,
      "eval_samples_per_second": 5.317,
      "eval_steps_per_second": 1.427,
      "step": 1215
    },
    {
      "epoch": 15.0,
      "learning_rate": 1.697530864197531e-06,
      "loss": 0.1911,
      "step": 1220
    },
    {
      "epoch": 15.01,
      "learning_rate": 1.6203703703703705e-06,
      "loss": 0.0041,
      "step": 1230
    },
    {
      "epoch": 15.02,
      "learning_rate": 1.54320987654321e-06,
      "loss": 0.3306,
      "step": 1240
    },
    {
      "epoch": 15.02,
      "learning_rate": 1.4660493827160497e-06,
      "loss": 0.1108,
      "step": 1250
    },
    {
      "epoch": 15.03,
      "learning_rate": 1.3888888888888892e-06,
      "loss": 0.1808,
      "step": 1260
    },
    {
      "epoch": 15.04,
      "learning_rate": 1.3117283950617284e-06,
      "loss": 0.1984,
      "step": 1270
    },
    {
      "epoch": 15.05,
      "learning_rate": 1.234567901234568e-06,
      "loss": 0.0157,
      "step": 1280
    },
    {
      "epoch": 15.05,
      "learning_rate": 1.1574074074074076e-06,
      "loss": 0.4692,
      "step": 1290
    },
    {
      "epoch": 15.06,
      "eval_accuracy": 0.7317073170731707,
      "eval_loss": 1.0264424085617065,
      "eval_runtime": 7.57,
      "eval_samples_per_second": 5.416,
      "eval_steps_per_second": 1.453,
      "step": 1296
    },
    {
      "epoch": 15.06,
      "step": 1296,
      "total_flos": 2.262256757640895e+19,
      "train_loss": 0.4041450611419148,
      "train_runtime": 2319.3656,
      "train_samples_per_second": 2.483,
      "train_steps_per_second": 0.621
    },
    {
      "epoch": 15.06,
      "eval_accuracy": 0.8292682926829268,
      "eval_loss": 0.4281896650791168,
      "eval_runtime": 7.8902,
      "eval_samples_per_second": 5.196,
      "eval_steps_per_second": 1.394,
      "step": 1296
    },
    {
      "epoch": 15.06,
      "eval_accuracy": 0.8292682926829268,
      "eval_loss": 0.4281897246837616,
      "eval_runtime": 7.7423,
      "eval_samples_per_second": 5.296,
      "eval_steps_per_second": 1.421,
      "step": 1296
    }
  ],
  "logging_steps": 10,
  "max_steps": 1440,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 2.262256757640895e+19,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}