{
  "best_metric": 0.728110599078341,
  "best_model_checkpoint": "videomae-base-finetuned-subset-check10/checkpoint-1960",
  "epoch": 39.016216216216215,
  "eval_steps": 500,
  "global_step": 2220,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 4.504504504504505e-07,
      "loss": 1.5494,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.00900900900901e-07,
      "loss": 1.5329,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.3513513513513515e-06,
      "loss": 1.5277,
      "step": 30
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.801801801801802e-06,
      "loss": 1.5543,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.2522522522522524e-06,
      "loss": 1.5175,
      "step": 50
    },
    {
      "epoch": 0.03,
      "eval_accuracy": 0.2073732718894009,
      "eval_loss": 1.6040809154510498,
      "eval_runtime": 183.7832,
      "eval_samples_per_second": 1.181,
      "eval_steps_per_second": 0.152,
      "step": 56
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.702702702702703e-06,
      "loss": 1.4668,
      "step": 60
    },
    {
      "epoch": 1.01,
      "learning_rate": 3.1531531531531532e-06,
      "loss": 1.4494,
      "step": 70
    },
    {
      "epoch": 1.01,
      "learning_rate": 3.603603603603604e-06,
      "loss": 1.5305,
      "step": 80
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.0540540540540545e-06,
      "loss": 1.4669,
      "step": 90
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.504504504504505e-06,
      "loss": 1.4467,
      "step": 100
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.954954954954955e-06,
      "loss": 1.4397,
      "step": 110
    },
    {
      "epoch": 1.03,
      "eval_accuracy": 0.3870967741935484,
      "eval_loss": 1.4559013843536377,
      "eval_runtime": 165.5917,
      "eval_samples_per_second": 1.31,
      "eval_steps_per_second": 0.169,
      "step": 112
    },
    {
      "epoch": 2.0,
      "learning_rate": 5.405405405405406e-06,
      "loss": 1.3919,
      "step": 120
    },
    {
      "epoch": 2.01,
      "learning_rate": 5.855855855855856e-06,
      "loss": 1.4257,
      "step": 130
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.3063063063063065e-06,
      "loss": 1.402,
      "step": 140
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.7567567567567575e-06,
      "loss": 1.3609,
      "step": 150
    },
    {
      "epoch": 2.02,
      "learning_rate": 7.207207207207208e-06,
      "loss": 1.464,
      "step": 160
    },
    {
      "epoch": 2.03,
      "eval_accuracy": 0.39631336405529954,
      "eval_loss": 1.3637099266052246,
      "eval_runtime": 165.2932,
      "eval_samples_per_second": 1.313,
      "eval_steps_per_second": 0.169,
      "step": 168
    },
    {
      "epoch": 3.0,
      "learning_rate": 7.657657657657658e-06,
      "loss": 1.3384,
      "step": 170
    },
    {
      "epoch": 3.01,
      "learning_rate": 8.108108108108109e-06,
      "loss": 1.4317,
      "step": 180
    },
    {
      "epoch": 3.01,
      "learning_rate": 8.55855855855856e-06,
      "loss": 1.1669,
      "step": 190
    },
    {
      "epoch": 3.01,
      "learning_rate": 9.00900900900901e-06,
      "loss": 1.4449,
      "step": 200
    },
    {
      "epoch": 3.02,
      "learning_rate": 9.45945945945946e-06,
      "loss": 1.3429,
      "step": 210
    },
    {
      "epoch": 3.02,
      "learning_rate": 9.90990990990991e-06,
      "loss": 1.3404,
      "step": 220
    },
    {
      "epoch": 3.03,
      "eval_accuracy": 0.4470046082949309,
      "eval_loss": 1.2467495203018188,
      "eval_runtime": 166.7691,
      "eval_samples_per_second": 1.301,
      "eval_steps_per_second": 0.168,
      "step": 224
    },
    {
      "epoch": 4.0,
      "learning_rate": 9.95995995995996e-06,
      "loss": 1.2682,
      "step": 230
    },
    {
      "epoch": 4.01,
      "learning_rate": 9.90990990990991e-06,
      "loss": 1.2307,
      "step": 240
    },
    {
      "epoch": 4.01,
      "learning_rate": 9.85985985985986e-06,
      "loss": 1.14,
      "step": 250
    },
    {
      "epoch": 4.02,
      "learning_rate": 9.80980980980981e-06,
      "loss": 1.091,
      "step": 260
    },
    {
      "epoch": 4.02,
      "learning_rate": 9.75975975975976e-06,
      "loss": 1.1822,
      "step": 270
    },
    {
      "epoch": 4.03,
      "learning_rate": 9.70970970970971e-06,
      "loss": 1.3284,
      "step": 280
    },
    {
      "epoch": 4.03,
      "eval_accuracy": 0.3317972350230415,
      "eval_loss": 1.3115123510360718,
      "eval_runtime": 165.1237,
      "eval_samples_per_second": 1.314,
      "eval_steps_per_second": 0.17,
      "step": 280
    },
    {
      "epoch": 5.0,
      "learning_rate": 9.65965965965966e-06,
      "loss": 1.1944,
      "step": 290
    },
    {
      "epoch": 5.01,
      "learning_rate": 9.60960960960961e-06,
      "loss": 1.3055,
      "step": 300
    },
    {
      "epoch": 5.01,
      "learning_rate": 9.55955955955956e-06,
      "loss": 1.0722,
      "step": 310
    },
    {
      "epoch": 5.02,
      "learning_rate": 9.50950950950951e-06,
      "loss": 1.0467,
      "step": 320
    },
    {
      "epoch": 5.02,
      "learning_rate": 9.45945945945946e-06,
      "loss": 1.1598,
      "step": 330
    },
    {
      "epoch": 5.03,
      "eval_accuracy": 0.4470046082949309,
      "eval_loss": 1.2488903999328613,
      "eval_runtime": 168.3906,
      "eval_samples_per_second": 1.289,
      "eval_steps_per_second": 0.166,
      "step": 336
    },
    {
      "epoch": 6.0,
      "learning_rate": 9.40940940940941e-06,
      "loss": 1.2285,
      "step": 340
    },
    {
      "epoch": 6.01,
      "learning_rate": 9.35935935935936e-06,
      "loss": 1.0002,
      "step": 350
    },
    {
      "epoch": 6.01,
      "learning_rate": 9.30930930930931e-06,
      "loss": 1.2506,
      "step": 360
    },
    {
      "epoch": 6.02,
      "learning_rate": 9.25925925925926e-06,
      "loss": 1.0963,
      "step": 370
    },
    {
      "epoch": 6.02,
      "learning_rate": 9.20920920920921e-06,
      "loss": 1.1495,
      "step": 380
    },
    {
      "epoch": 6.02,
      "learning_rate": 9.15915915915916e-06,
      "loss": 0.9615,
      "step": 390
    },
    {
      "epoch": 6.03,
      "eval_accuracy": 0.4009216589861751,
      "eval_loss": 1.3057225942611694,
      "eval_runtime": 165.4477,
      "eval_samples_per_second": 1.312,
      "eval_steps_per_second": 0.169,
      "step": 392
    },
    {
      "epoch": 7.0,
      "learning_rate": 9.10910910910911e-06,
      "loss": 1.0211,
      "step": 400
    },
    {
      "epoch": 7.01,
      "learning_rate": 9.05905905905906e-06,
      "loss": 1.0622,
      "step": 410
    },
    {
      "epoch": 7.01,
      "learning_rate": 9.00900900900901e-06,
      "loss": 1.1759,
      "step": 420
    },
    {
      "epoch": 7.02,
      "learning_rate": 8.95895895895896e-06,
      "loss": 1.0414,
      "step": 430
    },
    {
      "epoch": 7.02,
      "learning_rate": 8.90890890890891e-06,
      "loss": 0.9357,
      "step": 440
    },
    {
      "epoch": 7.03,
      "eval_accuracy": 0.6497695852534562,
      "eval_loss": 0.9200783967971802,
      "eval_runtime": 165.3979,
      "eval_samples_per_second": 1.312,
      "eval_steps_per_second": 0.169,
      "step": 448
    },
    {
      "epoch": 8.0,
      "learning_rate": 8.85885885885886e-06,
      "loss": 1.2506,
      "step": 450
    },
    {
      "epoch": 8.01,
      "learning_rate": 8.80880880880881e-06,
      "loss": 1.0342,
      "step": 460
    },
    {
      "epoch": 8.01,
      "learning_rate": 8.75875875875876e-06,
      "loss": 1.0815,
      "step": 470
    },
    {
      "epoch": 8.01,
      "learning_rate": 8.70870870870871e-06,
      "loss": 1.0373,
      "step": 480
    },
    {
      "epoch": 8.02,
      "learning_rate": 8.65865865865866e-06,
      "loss": 0.9051,
      "step": 490
    },
    {
      "epoch": 8.02,
      "learning_rate": 8.60860860860861e-06,
      "loss": 0.9785,
      "step": 500
    },
    {
      "epoch": 8.03,
      "eval_accuracy": 0.6774193548387096,
      "eval_loss": 0.8629115223884583,
      "eval_runtime": 165.2531,
      "eval_samples_per_second": 1.313,
      "eval_steps_per_second": 0.169,
      "step": 504
    },
    {
      "epoch": 9.0,
      "learning_rate": 8.55855855855856e-06,
      "loss": 0.9915,
      "step": 510
    },
    {
      "epoch": 9.01,
      "learning_rate": 8.50850850850851e-06,
      "loss": 0.9176,
      "step": 520
    },
    {
      "epoch": 9.01,
      "learning_rate": 8.45845845845846e-06,
      "loss": 1.1561,
      "step": 530
    },
    {
      "epoch": 9.02,
      "learning_rate": 8.408408408408409e-06,
      "loss": 0.8966,
      "step": 540
    },
    {
      "epoch": 9.02,
      "learning_rate": 8.358358358358359e-06,
      "loss": 0.894,
      "step": 550
    },
    {
      "epoch": 9.03,
      "learning_rate": 8.308308308308309e-06,
      "loss": 1.0862,
      "step": 560
    },
    {
      "epoch": 9.03,
      "eval_accuracy": 0.5069124423963134,
      "eval_loss": 1.0976766347885132,
      "eval_runtime": 165.1522,
      "eval_samples_per_second": 1.314,
      "eval_steps_per_second": 0.17,
      "step": 560
    },
    {
      "epoch": 10.0,
      "learning_rate": 8.258258258258259e-06,
      "loss": 0.7806,
      "step": 570
    },
    {
      "epoch": 10.01,
      "learning_rate": 8.208208208208209e-06,
      "loss": 0.9904,
      "step": 580
    },
    {
      "epoch": 10.01,
      "learning_rate": 8.158158158158159e-06,
      "loss": 0.9099,
      "step": 590
    },
    {
      "epoch": 10.02,
      "learning_rate": 8.108108108108109e-06,
      "loss": 0.9169,
      "step": 600
    },
    {
      "epoch": 10.02,
      "learning_rate": 8.058058058058059e-06,
      "loss": 0.9315,
      "step": 610
    },
    {
      "epoch": 10.03,
      "eval_accuracy": 0.7096774193548387,
      "eval_loss": 0.7867635488510132,
      "eval_runtime": 165.0158,
      "eval_samples_per_second": 1.315,
      "eval_steps_per_second": 0.17,
      "step": 616
    },
    {
      "epoch": 11.0,
      "learning_rate": 8.00800800800801e-06,
      "loss": 0.7636,
      "step": 620
    },
    {
      "epoch": 11.01,
      "learning_rate": 7.95795795795796e-06,
      "loss": 0.9198,
      "step": 630
    },
    {
      "epoch": 11.01,
      "learning_rate": 7.90790790790791e-06,
      "loss": 1.0816,
      "step": 640
    },
    {
      "epoch": 11.02,
      "learning_rate": 7.85785785785786e-06,
      "loss": 0.8409,
      "step": 650
    },
    {
      "epoch": 11.02,
      "learning_rate": 7.807807807807808e-06,
      "loss": 0.7348,
      "step": 660
    },
    {
      "epoch": 11.02,
      "learning_rate": 7.757757757757758e-06,
      "loss": 0.9404,
      "step": 670
    },
    {
      "epoch": 11.03,
      "eval_accuracy": 0.6728110599078341,
      "eval_loss": 0.8170290589332581,
      "eval_runtime": 165.2251,
      "eval_samples_per_second": 1.313,
      "eval_steps_per_second": 0.169,
      "step": 672
    },
    {
      "epoch": 12.0,
      "learning_rate": 7.707707707707708e-06,
      "loss": 0.8669,
      "step": 680
    },
    {
      "epoch": 12.01,
      "learning_rate": 7.657657657657658e-06,
      "loss": 0.7773,
      "step": 690
    },
    {
      "epoch": 12.01,
      "learning_rate": 7.607607607607608e-06,
      "loss": 0.7599,
      "step": 700
    },
    {
      "epoch": 12.02,
      "learning_rate": 7.557557557557558e-06,
      "loss": 0.9216,
      "step": 710
    },
    {
      "epoch": 12.02,
      "learning_rate": 7.507507507507507e-06,
      "loss": 0.939,
      "step": 720
    },
    {
      "epoch": 12.03,
      "eval_accuracy": 0.663594470046083,
      "eval_loss": 0.9246166348457336,
      "eval_runtime": 165.7726,
      "eval_samples_per_second": 1.309,
      "eval_steps_per_second": 0.169,
      "step": 728
    },
    {
      "epoch": 13.0,
      "learning_rate": 7.457457457457457e-06,
      "loss": 0.9703,
      "step": 730
    },
    {
      "epoch": 13.01,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.6688,
      "step": 740
    },
    {
      "epoch": 13.01,
      "learning_rate": 7.3573573573573575e-06,
      "loss": 0.9097,
      "step": 750
    },
    {
      "epoch": 13.01,
      "learning_rate": 7.307307307307308e-06,
      "loss": 0.9963,
      "step": 760
    },
    {
      "epoch": 13.02,
      "learning_rate": 7.257257257257258e-06,
      "loss": 0.8594,
      "step": 770
    },
    {
      "epoch": 13.02,
      "learning_rate": 7.207207207207208e-06,
      "loss": 0.8205,
      "step": 780
    },
    {
      "epoch": 13.03,
      "eval_accuracy": 0.6866359447004609,
      "eval_loss": 0.8420272469520569,
      "eval_runtime": 165.0282,
      "eval_samples_per_second": 1.315,
      "eval_steps_per_second": 0.17,
      "step": 784
    },
    {
      "epoch": 14.0,
      "learning_rate": 7.157157157157158e-06,
      "loss": 0.8206,
      "step": 790
    },
    {
      "epoch": 14.01,
      "learning_rate": 7.107107107107107e-06,
      "loss": 0.8183,
      "step": 800
    },
    {
      "epoch": 14.01,
      "learning_rate": 7.057057057057057e-06,
      "loss": 0.7596,
      "step": 810
    },
    {
      "epoch": 14.02,
      "learning_rate": 7.007007007007007e-06,
      "loss": 0.799,
      "step": 820
    },
    {
      "epoch": 14.02,
      "learning_rate": 6.956956956956957e-06,
      "loss": 0.7377,
      "step": 830
    },
    {
      "epoch": 14.03,
      "learning_rate": 6.906906906906907e-06,
      "loss": 0.6719,
      "step": 840
    },
    {
      "epoch": 14.03,
      "eval_accuracy": 0.5898617511520737,
      "eval_loss": 1.07249116897583,
      "eval_runtime": 164.1063,
      "eval_samples_per_second": 1.322,
      "eval_steps_per_second": 0.171,
      "step": 840
    },
    {
      "epoch": 15.0,
      "learning_rate": 6.856856856856857e-06,
      "loss": 0.6874,
      "step": 850
    },
    {
      "epoch": 15.01,
      "learning_rate": 6.8068068068068075e-06,
      "loss": 0.8207,
      "step": 860
    },
    {
      "epoch": 15.01,
      "learning_rate": 6.7567567567567575e-06,
      "loss": 0.7454,
      "step": 870
    },
    {
      "epoch": 15.02,
      "learning_rate": 6.706706706706707e-06,
      "loss": 0.7188,
      "step": 880
    },
    {
      "epoch": 15.02,
      "learning_rate": 6.656656656656657e-06,
      "loss": 0.8308,
      "step": 890
    },
    {
      "epoch": 15.03,
      "eval_accuracy": 0.6912442396313364,
      "eval_loss": 0.8682537078857422,
      "eval_runtime": 165.7327,
      "eval_samples_per_second": 1.309,
      "eval_steps_per_second": 0.169,
      "step": 896
    },
    {
      "epoch": 16.0,
      "learning_rate": 6.606606606606607e-06,
      "loss": 0.891,
      "step": 900
    },
    {
      "epoch": 16.01,
      "learning_rate": 6.556556556556557e-06,
      "loss": 0.7597,
      "step": 910
    },
    {
      "epoch": 16.01,
      "learning_rate": 6.506506506506507e-06,
      "loss": 0.8318,
      "step": 920
    },
    {
      "epoch": 16.02,
      "learning_rate": 6.456456456456457e-06,
      "loss": 0.7681,
      "step": 930
    },
    {
      "epoch": 16.02,
      "learning_rate": 6.406406406406407e-06,
      "loss": 1.0094,
      "step": 940
    },
    {
      "epoch": 16.02,
      "learning_rate": 6.356356356356357e-06,
      "loss": 0.7554,
      "step": 950
    },
    {
      "epoch": 16.03,
      "eval_accuracy": 0.5990783410138248,
      "eval_loss": 0.9684067368507385,
      "eval_runtime": 164.5634,
      "eval_samples_per_second": 1.319,
      "eval_steps_per_second": 0.17,
      "step": 952
    },
    {
      "epoch": 17.0,
      "learning_rate": 6.3063063063063065e-06,
      "loss": 0.9087,
      "step": 960
    },
    {
      "epoch": 17.01,
      "learning_rate": 6.2562562562562565e-06,
      "loss": 0.7401,
      "step": 970
    },
    {
      "epoch": 17.01,
      "learning_rate": 6.206206206206207e-06,
      "loss": 0.7611,
      "step": 980
    },
    {
      "epoch": 17.02,
      "learning_rate": 6.156156156156157e-06,
      "loss": 0.6347,
      "step": 990
    },
    {
      "epoch": 17.02,
      "learning_rate": 6.106106106106107e-06,
      "loss": 0.6962,
      "step": 1000
    },
    {
      "epoch": 17.03,
      "eval_accuracy": 0.5483870967741935,
      "eval_loss": 1.1106446981430054,
      "eval_runtime": 165.0022,
      "eval_samples_per_second": 1.315,
      "eval_steps_per_second": 0.17,
      "step": 1008
    },
    {
      "epoch": 18.0,
      "learning_rate": 6.056056056056057e-06,
      "loss": 0.6861,
      "step": 1010
    },
    {
      "epoch": 18.01,
      "learning_rate": 6.006006006006007e-06,
      "loss": 0.7999,
      "step": 1020
    },
    {
      "epoch": 18.01,
      "learning_rate": 5.955955955955957e-06,
      "loss": 0.7412,
      "step": 1030
    },
    {
      "epoch": 18.01,
      "learning_rate": 5.905905905905906e-06,
      "loss": 0.8145,
      "step": 1040
    },
    {
      "epoch": 18.02,
      "learning_rate": 5.855855855855856e-06,
      "loss": 0.6653,
      "step": 1050
    },
    {
      "epoch": 18.02,
      "learning_rate": 5.805805805805806e-06,
      "loss": 0.7995,
      "step": 1060
    },
    {
      "epoch": 18.03,
      "eval_accuracy": 0.6497695852534562,
      "eval_loss": 0.9750950336456299,
      "eval_runtime": 163.7137,
      "eval_samples_per_second": 1.325,
      "eval_steps_per_second": 0.171,
      "step": 1064
    },
    {
      "epoch": 19.0,
      "learning_rate": 5.755755755755756e-06,
      "loss": 0.6923,
      "step": 1070
    },
    {
      "epoch": 19.01,
      "learning_rate": 5.7057057057057065e-06,
      "loss": 0.7874,
      "step": 1080
    },
    {
      "epoch": 19.01,
      "learning_rate": 5.6556556556556565e-06,
      "loss": 0.728,
      "step": 1090
    },
    {
      "epoch": 19.02,
      "learning_rate": 5.605605605605607e-06,
      "loss": 0.6329,
      "step": 1100
    },
    {
      "epoch": 19.02,
      "learning_rate": 5.555555555555557e-06,
      "loss": 0.8939,
      "step": 1110
    },
    {
      "epoch": 19.03,
      "learning_rate": 5.505505505505506e-06,
      "loss": 0.8298,
      "step": 1120
    },
    {
      "epoch": 19.03,
      "eval_accuracy": 0.5299539170506913,
      "eval_loss": 1.0630956888198853,
      "eval_runtime": 165.0711,
      "eval_samples_per_second": 1.315,
      "eval_steps_per_second": 0.17,
      "step": 1120
    },
    {
      "epoch": 20.0,
      "learning_rate": 5.455455455455456e-06,
      "loss": 0.7025,
      "step": 1130
    },
    {
      "epoch": 20.01,
      "learning_rate": 5.405405405405406e-06,
      "loss": 0.6251,
      "step": 1140
    },
    {
      "epoch": 20.01,
      "learning_rate": 5.355355355355356e-06,
      "loss": 0.7818,
      "step": 1150
    },
    {
      "epoch": 20.02,
      "learning_rate": 5.305305305305306e-06,
      "loss": 0.8259,
      "step": 1160
    },
    {
      "epoch": 20.02,
      "learning_rate": 5.255255255255256e-06,
      "loss": 0.6607,
      "step": 1170
    },
    {
      "epoch": 20.03,
      "eval_accuracy": 0.6175115207373272,
      "eval_loss": 0.9457883834838867,
      "eval_runtime": 164.6389,
      "eval_samples_per_second": 1.318,
      "eval_steps_per_second": 0.17,
      "step": 1176
    },
    {
      "epoch": 21.0,
      "learning_rate": 5.205205205205206e-06,
      "loss": 0.6344,
      "step": 1180
    },
    {
      "epoch": 21.01,
      "learning_rate": 5.155155155155156e-06,
      "loss": 0.637,
      "step": 1190
    },
    {
      "epoch": 21.01,
      "learning_rate": 5.105105105105106e-06,
      "loss": 0.858,
      "step": 1200
    },
    {
      "epoch": 21.02,
      "learning_rate": 5.055055055055056e-06,
      "loss": 0.6934,
      "step": 1210
    },
    {
      "epoch": 21.02,
      "learning_rate": 5.005005005005006e-06,
      "loss": 0.5714,
      "step": 1220
    },
    {
      "epoch": 21.02,
      "learning_rate": 4.954954954954955e-06,
      "loss": 0.688,
      "step": 1230
    },
    {
      "epoch": 21.03,
      "eval_accuracy": 0.6036866359447005,
      "eval_loss": 1.029563546180725,
      "eval_runtime": 163.856,
      "eval_samples_per_second": 1.324,
      "eval_steps_per_second": 0.171,
      "step": 1232
    },
    {
      "epoch": 22.0,
      "learning_rate": 4.904904904904905e-06,
      "loss": 0.5114,
      "step": 1240
    },
    {
      "epoch": 22.01,
      "learning_rate": 4.854854854854855e-06,
      "loss": 0.6638,
      "step": 1250
    },
    {
      "epoch": 22.01,
      "learning_rate": 4.804804804804805e-06,
      "loss": 0.7833,
      "step": 1260
    },
    {
      "epoch": 22.02,
      "learning_rate": 4.754754754754755e-06,
      "loss": 0.7556,
      "step": 1270
    },
    {
      "epoch": 22.02,
      "learning_rate": 4.704704704704705e-06,
      "loss": 0.5835,
      "step": 1280
    },
    {
      "epoch": 22.03,
      "eval_accuracy": 0.6774193548387096,
      "eval_loss": 0.8947747349739075,
      "eval_runtime": 163.875,
      "eval_samples_per_second": 1.324,
      "eval_steps_per_second": 0.171,
      "step": 1288
    },
    {
      "epoch": 23.0,
      "learning_rate": 4.654654654654655e-06,
      "loss": 0.7209,
      "step": 1290
    },
    {
      "epoch": 23.01,
      "learning_rate": 4.604604604604605e-06,
      "loss": 0.6087,
      "step": 1300
    },
    {
      "epoch": 23.01,
      "learning_rate": 4.554554554554555e-06,
      "loss": 0.6529,
      "step": 1310
    },
    {
      "epoch": 23.01,
      "learning_rate": 4.504504504504505e-06,
      "loss": 0.7221,
      "step": 1320
    },
    {
      "epoch": 23.02,
      "learning_rate": 4.454454454454455e-06,
      "loss": 0.6167,
      "step": 1330
    },
    {
      "epoch": 23.02,
      "learning_rate": 4.404404404404405e-06,
      "loss": 0.6987,
      "step": 1340
    },
    {
      "epoch": 23.03,
      "eval_accuracy": 0.7188940092165899,
      "eval_loss": 0.7882533669471741,
      "eval_runtime": 163.4201,
      "eval_samples_per_second": 1.328,
      "eval_steps_per_second": 0.171,
      "step": 1344
    },
    {
      "epoch": 24.0,
      "learning_rate": 4.354354354354355e-06,
      "loss": 0.7234,
      "step": 1350
    },
    {
      "epoch": 24.01,
      "learning_rate": 4.304304304304305e-06,
      "loss": 0.808,
      "step": 1360
    },
    {
      "epoch": 24.01,
      "learning_rate": 4.254254254254255e-06,
      "loss": 0.6323,
      "step": 1370
    },
    {
      "epoch": 24.02,
      "learning_rate": 4.204204204204204e-06,
      "loss": 0.5463,
      "step": 1380
    },
    {
      "epoch": 24.02,
      "learning_rate": 4.154154154154154e-06,
      "loss": 0.7978,
      "step": 1390
    },
    {
      "epoch": 24.03,
      "learning_rate": 4.1041041041041045e-06,
      "loss": 0.4979,
      "step": 1400
    },
    {
      "epoch": 24.03,
      "eval_accuracy": 0.7188940092165899,
      "eval_loss": 0.7089133262634277,
      "eval_runtime": 164.3791,
      "eval_samples_per_second": 1.32,
      "eval_steps_per_second": 0.17,
      "step": 1400
    },
    {
      "epoch": 25.0,
      "learning_rate": 4.0540540540540545e-06,
      "loss": 0.4685,
      "step": 1410
    },
    {
      "epoch": 25.01,
      "learning_rate": 4.004004004004005e-06,
      "loss": 0.665,
      "step": 1420
    },
    {
      "epoch": 25.01,
      "learning_rate": 3.953953953953955e-06,
      "loss": 0.6872,
      "step": 1430
    },
    {
      "epoch": 25.02,
      "learning_rate": 3.903903903903904e-06,
      "loss": 0.5854,
      "step": 1440
    },
    {
      "epoch": 25.02,
      "learning_rate": 3.853853853853854e-06,
      "loss": 0.6163,
      "step": 1450
    },
    {
      "epoch": 25.03,
      "eval_accuracy": 0.7235023041474654,
      "eval_loss": 0.7633541226387024,
      "eval_runtime": 163.5735,
      "eval_samples_per_second": 1.327,
      "eval_steps_per_second": 0.171,
      "step": 1456
    },
    {
      "epoch": 26.0,
      "learning_rate": 3.803803803803804e-06,
      "loss": 0.7065,
      "step": 1460
    },
    {
      "epoch": 26.01,
      "learning_rate": 3.7537537537537537e-06,
      "loss": 0.6684,
      "step": 1470
    },
    {
      "epoch": 26.01,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.786,
      "step": 1480
    },
    {
      "epoch": 26.02,
      "learning_rate": 3.653653653653654e-06,
      "loss": 0.437,
      "step": 1490
    },
    {
      "epoch": 26.02,
      "learning_rate": 3.603603603603604e-06,
      "loss": 0.6708,
      "step": 1500
    },
    {
      "epoch": 26.02,
      "learning_rate": 3.5535535535535535e-06,
      "loss": 0.6754,
      "step": 1510
    },
    {
      "epoch": 26.03,
      "eval_accuracy": 0.6359447004608295,
      "eval_loss": 0.9443588852882385,
      "eval_runtime": 163.9867,
      "eval_samples_per_second": 1.323,
      "eval_steps_per_second": 0.171,
      "step": 1512
    },
    {
      "epoch": 27.0,
      "learning_rate": 3.5035035035035036e-06,
      "loss": 0.7284,
      "step": 1520
    },
    {
      "epoch": 27.01,
      "learning_rate": 3.4534534534534537e-06,
      "loss": 0.5413,
      "step": 1530
    },
    {
      "epoch": 27.01,
      "learning_rate": 3.4034034034034037e-06,
      "loss": 0.6884,
      "step": 1540
    },
    {
      "epoch": 27.02,
      "learning_rate": 3.3533533533533534e-06,
      "loss": 0.5541,
      "step": 1550
    },
    {
      "epoch": 27.02,
      "learning_rate": 3.3033033033033035e-06,
      "loss": 0.6673,
      "step": 1560
    },
    {
      "epoch": 27.03,
      "eval_accuracy": 0.6543778801843319,
      "eval_loss": 0.8390823602676392,
      "eval_runtime": 164.572,
      "eval_samples_per_second": 1.319,
      "eval_steps_per_second": 0.17,
      "step": 1568
    },
    {
      "epoch": 28.0,
      "learning_rate": 3.2532532532532535e-06,
      "loss": 0.6895,
      "step": 1570
    },
    {
      "epoch": 28.01,
      "learning_rate": 3.2032032032032036e-06,
      "loss": 0.553,
      "step": 1580
    },
    {
      "epoch": 28.01,
      "learning_rate": 3.1531531531531532e-06,
      "loss": 0.4311,
      "step": 1590
    },
    {
      "epoch": 28.01,
      "learning_rate": 3.1031031031031033e-06,
      "loss": 0.6969,
      "step": 1600
    },
    {
      "epoch": 28.02,
      "learning_rate": 3.0530530530530534e-06,
      "loss": 0.7507,
      "step": 1610
    },
    {
      "epoch": 28.02,
      "learning_rate": 3.0030030030030034e-06,
      "loss": 0.4924,
      "step": 1620
    },
    {
      "epoch": 28.03,
      "eval_accuracy": 0.6682027649769585,
      "eval_loss": 0.8288503885269165,
      "eval_runtime": 164.6437,
      "eval_samples_per_second": 1.318,
      "eval_steps_per_second": 0.17,
      "step": 1624
    },
    {
      "epoch": 29.0,
      "learning_rate": 2.952952952952953e-06,
      "loss": 0.7695,
      "step": 1630
    },
    {
      "epoch": 29.01,
      "learning_rate": 2.902902902902903e-06,
      "loss": 0.5981,
      "step": 1640
    },
    {
      "epoch": 29.01,
      "learning_rate": 2.8528528528528532e-06,
      "loss": 0.613,
      "step": 1650
    },
    {
      "epoch": 29.02,
      "learning_rate": 2.8028028028028033e-06,
      "loss": 0.5841,
      "step": 1660
    },
    {
      "epoch": 29.02,
      "learning_rate": 2.752752752752753e-06,
      "loss": 0.7704,
      "step": 1670
    },
    {
      "epoch": 29.03,
      "learning_rate": 2.702702702702703e-06,
      "loss": 0.6438,
      "step": 1680
    },
    {
      "epoch": 29.03,
      "eval_accuracy": 0.6129032258064516,
      "eval_loss": 0.9605286121368408,
      "eval_runtime": 164.5062,
      "eval_samples_per_second": 1.319,
      "eval_steps_per_second": 0.17,
      "step": 1680
    },
    {
      "epoch": 30.0,
      "learning_rate": 2.652652652652653e-06,
      "loss": 0.583,
      "step": 1690
    },
    {
      "epoch": 30.01,
      "learning_rate": 2.602602602602603e-06,
      "loss": 0.7539,
      "step": 1700
    },
    {
      "epoch": 30.01,
      "learning_rate": 2.552552552552553e-06,
      "loss": 0.5418,
      "step": 1710
    },
    {
      "epoch": 30.02,
      "learning_rate": 2.502502502502503e-06,
      "loss": 0.6689,
      "step": 1720
    },
    {
      "epoch": 30.02,
      "learning_rate": 2.4524524524524525e-06,
      "loss": 0.5714,
      "step": 1730
    },
    {
      "epoch": 30.03,
      "eval_accuracy": 0.6451612903225806,
      "eval_loss": 0.8838080763816833,
      "eval_runtime": 163.4686,
      "eval_samples_per_second": 1.327,
      "eval_steps_per_second": 0.171,
      "step": 1736
    },
    {
      "epoch": 31.0,
      "learning_rate": 2.4024024024024026e-06,
      "loss": 0.4737,
      "step": 1740
    },
    {
      "epoch": 31.01,
      "learning_rate": 2.3523523523523527e-06,
      "loss": 0.6563,
      "step": 1750
    },
    {
      "epoch": 31.01,
      "learning_rate": 2.3023023023023023e-06,
      "loss": 0.4469,
      "step": 1760
    },
    {
      "epoch": 31.02,
      "learning_rate": 2.2522522522522524e-06,
      "loss": 0.5655,
      "step": 1770
    },
    {
      "epoch": 31.02,
      "learning_rate": 2.2022022022022024e-06,
      "loss": 0.5421,
      "step": 1780
    },
    {
      "epoch": 31.02,
      "learning_rate": 2.1521521521521525e-06,
      "loss": 0.6726,
      "step": 1790
    },
    {
      "epoch": 31.03,
      "eval_accuracy": 0.6589861751152074,
      "eval_loss": 0.8412278890609741,
      "eval_runtime": 164.9282,
      "eval_samples_per_second": 1.316,
      "eval_steps_per_second": 0.17,
      "step": 1792
    },
    {
      "epoch": 32.0,
      "learning_rate": 2.102102102102102e-06,
      "loss": 0.7254,
      "step": 1800
    },
    {
      "epoch": 32.01,
      "learning_rate": 2.0520520520520522e-06,
      "loss": 0.5397,
      "step": 1810
    },
    {
      "epoch": 32.01,
      "learning_rate": 2.0020020020020023e-06,
      "loss": 0.596,
      "step": 1820
    },
    {
      "epoch": 32.02,
      "learning_rate": 1.951951951951952e-06,
      "loss": 0.5603,
      "step": 1830
    },
    {
      "epoch": 32.02,
      "learning_rate": 1.901901901901902e-06,
      "loss": 0.5027,
      "step": 1840
    },
    {
      "epoch": 32.03,
      "eval_accuracy": 0.6728110599078341,
      "eval_loss": 0.8439391255378723,
      "eval_runtime": 168.1229,
      "eval_samples_per_second": 1.291,
      "eval_steps_per_second": 0.167,
      "step": 1848
    },
    {
      "epoch": 33.0,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.6769,
      "step": 1850
    },
    {
      "epoch": 33.01,
      "learning_rate": 1.801801801801802e-06,
      "loss": 0.565,
      "step": 1860
    },
    {
      "epoch": 33.01,
      "learning_rate": 1.7517517517517518e-06,
      "loss": 0.556,
      "step": 1870
    },
    {
      "epoch": 33.01,
      "learning_rate": 1.7017017017017019e-06,
      "loss": 0.4187,
      "step": 1880
    },
    {
      "epoch": 33.02,
      "learning_rate": 1.6516516516516517e-06,
      "loss": 0.4718,
      "step": 1890
    },
    {
      "epoch": 33.02,
      "learning_rate": 1.6016016016016018e-06,
      "loss": 0.4649,
      "step": 1900
    },
    {
      "epoch": 33.03,
      "eval_accuracy": 0.6267281105990783,
      "eval_loss": 0.9524617791175842,
      "eval_runtime": 169.6407,
      "eval_samples_per_second": 1.279,
      "eval_steps_per_second": 0.165,
      "step": 1904
    },
    {
      "epoch": 34.0,
      "learning_rate": 1.5515515515515517e-06,
      "loss": 0.4505,
      "step": 1910
    },
    {
      "epoch": 34.01,
      "learning_rate": 1.5015015015015017e-06,
      "loss": 0.5412,
      "step": 1920
    },
    {
      "epoch": 34.01,
      "learning_rate": 1.4514514514514516e-06,
      "loss": 0.5931,
      "step": 1930
    },
    {
      "epoch": 34.02,
      "learning_rate": 1.4014014014014016e-06,
      "loss": 0.4183,
      "step": 1940
    },
    {
      "epoch": 34.02,
      "learning_rate": 1.3513513513513515e-06,
      "loss": 0.6046,
      "step": 1950
    },
    {
      "epoch": 34.03,
      "learning_rate": 1.3013013013013016e-06,
      "loss": 0.6625,
      "step": 1960
    },
    {
      "epoch": 34.03,
      "eval_accuracy": 0.728110599078341,
      "eval_loss": 0.7850246429443359,
      "eval_runtime": 164.4991,
      "eval_samples_per_second": 1.319,
      "eval_steps_per_second": 0.17,
      "step": 1960
    },
    {
      "epoch": 35.0,
      "learning_rate": 1.2512512512512514e-06,
      "loss": 0.5967,
      "step": 1970
    },
    {
      "epoch": 35.01,
      "learning_rate": 1.2012012012012013e-06,
      "loss": 0.5503,
      "step": 1980
    },
    {
      "epoch": 35.01,
      "learning_rate": 1.1511511511511512e-06,
      "loss": 0.5159,
      "step": 1990
    },
    {
      "epoch": 35.02,
      "learning_rate": 1.1011011011011012e-06,
      "loss": 0.4377,
      "step": 2000
    },
    {
      "epoch": 35.02,
      "learning_rate": 1.051051051051051e-06,
      "loss": 0.5793,
      "step": 2010
    },
    {
      "epoch": 35.03,
      "eval_accuracy": 0.6728110599078341,
      "eval_loss": 0.8481199741363525,
      "eval_runtime": 164.3359,
      "eval_samples_per_second": 1.32,
      "eval_steps_per_second": 0.17,
      "step": 2016
    },
    {
      "epoch": 36.0,
      "learning_rate": 1.0010010010010011e-06,
      "loss": 0.7009,
      "step": 2020
    },
    {
      "epoch": 36.01,
      "learning_rate": 9.50950950950951e-07,
      "loss": 0.5902,
      "step": 2030
    },
    {
      "epoch": 36.01,
      "learning_rate": 9.00900900900901e-07,
      "loss": 0.5366,
      "step": 2040
    },
    {
      "epoch": 36.02,
      "learning_rate": 8.508508508508509e-07,
      "loss": 0.5119,
      "step": 2050
    },
    {
      "epoch": 36.02,
      "learning_rate": 8.008008008008009e-07,
      "loss": 0.4946,
      "step": 2060
    },
    {
      "epoch": 36.02,
      "learning_rate": 7.507507507507509e-07,
      "loss": 0.6411,
      "step": 2070
    },
    {
      "epoch": 36.03,
      "eval_accuracy": 0.6589861751152074,
      "eval_loss": 0.8842198252677917,
      "eval_runtime": 164.6218,
      "eval_samples_per_second": 1.318,
      "eval_steps_per_second": 0.17,
      "step": 2072
    },
    {
      "epoch": 37.0,
      "learning_rate": 7.007007007007008e-07,
      "loss": 0.5172,
      "step": 2080
    },
    {
      "epoch": 37.01,
      "learning_rate": 6.506506506506508e-07,
      "loss": 0.3986,
      "step": 2090
    },
    {
      "epoch": 37.01,
      "learning_rate": 6.006006006006006e-07,
      "loss": 0.4807,
      "step": 2100
    },
    {
      "epoch": 37.02,
      "learning_rate": 5.505505505505506e-07,
      "loss": 0.4681,
      "step": 2110
    },
    {
      "epoch": 37.02,
      "learning_rate": 5.005005005005006e-07,
      "loss": 0.6592,
      "step": 2120
    },
    {
      "epoch": 37.03,
      "eval_accuracy": 0.6912442396313364,
      "eval_loss": 0.802787184715271,
      "eval_runtime": 163.9059,
      "eval_samples_per_second": 1.324,
      "eval_steps_per_second": 0.171,
      "step": 2128
    },
    {
      "epoch": 38.0,
      "learning_rate": 4.504504504504505e-07,
      "loss": 0.5769,
      "step": 2130
    },
    {
      "epoch": 38.01,
      "learning_rate": 4.0040040040040045e-07,
      "loss": 0.6323,
      "step": 2140
    },
    {
      "epoch": 38.01,
      "learning_rate": 3.503503503503504e-07,
      "loss": 0.5319,
      "step": 2150
    },
    {
      "epoch": 38.01,
      "learning_rate": 3.003003003003003e-07,
      "loss": 0.6306,
      "step": 2160
    },
    {
      "epoch": 38.02,
      "learning_rate": 2.502502502502503e-07,
      "loss": 0.6192,
      "step": 2170
    },
    {
      "epoch": 38.02,
      "learning_rate": 2.0020020020020022e-07,
      "loss": 0.5524,
      "step": 2180
    },
    {
      "epoch": 38.03,
      "eval_accuracy": 0.6866359447004609,
      "eval_loss": 0.821592390537262,
      "eval_runtime": 165.1926,
      "eval_samples_per_second": 1.314,
      "eval_steps_per_second": 0.169,
      "step": 2184
    },
    {
      "epoch": 39.0,
      "learning_rate": 1.5015015015015016e-07,
      "loss": 0.5701,
      "step": 2190
    },
    {
      "epoch": 39.01,
      "learning_rate": 1.0010010010010011e-07,
      "loss": 0.5891,
      "step": 2200
    },
    {
      "epoch": 39.01,
      "learning_rate": 5.0050050050050056e-08,
      "loss": 0.5807,
      "step": 2210
    },
    {
      "epoch": 39.02,
      "learning_rate": 0.0,
      "loss": 0.5697,
      "step": 2220
    },
    {
      "epoch": 39.02,
      "eval_accuracy": 0.6774193548387096,
      "eval_loss": 0.833946943283081,
      "eval_runtime": 164.2083,
      "eval_samples_per_second": 1.321,
      "eval_steps_per_second": 0.171,
      "step": 2220
    },
    {
      "epoch": 39.02,
      "step": 2220,
      "total_flos": 2.208334746599719e+19,
      "train_loss": 0.8170982454274152,
      "train_runtime": 23121.8467,
      "train_samples_per_second": 0.768,
      "train_steps_per_second": 0.096
    },
    {
      "epoch": 39.02,
      "eval_accuracy": 0.6342592592592593,
      "eval_loss": 0.8682467341423035,
      "eval_runtime": 165.599,
      "eval_samples_per_second": 1.304,
      "eval_steps_per_second": 0.163,
      "step": 2220
    },
    {
      "epoch": 39.02,
      "eval_accuracy": 0.6342592592592593,
      "eval_loss": 0.8682467341423035,
      "eval_runtime": 165.9338,
      "eval_samples_per_second": 1.302,
      "eval_steps_per_second": 0.163,
      "step": 2220
    }
  ],
  "logging_steps": 10,
  "max_steps": 2220,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 2.208334746599719e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}