{
  "best_metric": 0.7327188940092166,
  "best_model_checkpoint": "videomae-base-finetuned-subset-check10/checkpoint-896",
  "epoch": 19.041441441441442,
  "eval_steps": 500,
  "global_step": 1110,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 9.00900900900901e-07,
      "loss": 1.6459,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.801801801801802e-06,
      "loss": 1.636,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.702702702702703e-06,
      "loss": 1.6904,
      "step": 30
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.603603603603604e-06,
      "loss": 1.6413,
      "step": 40
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.504504504504505e-06,
      "loss": 1.6285,
      "step": 50
    },
    {
      "epoch": 0.05,
      "eval_accuracy": 0.3686635944700461,
      "eval_loss": 1.6004022359848022,
      "eval_runtime": 173.9244,
      "eval_samples_per_second": 1.248,
      "eval_steps_per_second": 0.161,
      "step": 56
    },
    {
      "epoch": 1.0,
      "learning_rate": 5.405405405405406e-06,
      "loss": 1.63,
      "step": 60
    },
    {
      "epoch": 1.01,
      "learning_rate": 6.3063063063063065e-06,
      "loss": 1.5959,
      "step": 70
    },
    {
      "epoch": 1.02,
      "learning_rate": 7.207207207207208e-06,
      "loss": 1.6112,
      "step": 80
    },
    {
      "epoch": 1.03,
      "learning_rate": 8.108108108108109e-06,
      "loss": 1.6242,
      "step": 90
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.00900900900901e-06,
      "loss": 1.5879,
      "step": 100
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.90990990990991e-06,
      "loss": 1.6056,
      "step": 110
    },
    {
      "epoch": 1.05,
      "eval_accuracy": 0.19815668202764977,
      "eval_loss": 1.5825837850570679,
      "eval_runtime": 167.123,
      "eval_samples_per_second": 1.298,
      "eval_steps_per_second": 0.168,
      "step": 112
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.90990990990991e-06,
      "loss": 1.5988,
      "step": 120
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.80980980980981e-06,
      "loss": 1.5945,
      "step": 130
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.70970970970971e-06,
      "loss": 1.6349,
      "step": 140
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.60960960960961e-06,
      "loss": 1.5828,
      "step": 150
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.50950950950951e-06,
      "loss": 1.5254,
      "step": 160
    },
    {
      "epoch": 2.05,
      "eval_accuracy": 0.14746543778801843,
      "eval_loss": 1.5978792905807495,
      "eval_runtime": 166.6054,
      "eval_samples_per_second": 1.302,
      "eval_steps_per_second": 0.168,
      "step": 168
    },
    {
      "epoch": 3.0,
      "learning_rate": 9.40940940940941e-06,
      "loss": 1.5397,
      "step": 170
    },
    {
      "epoch": 3.01,
      "learning_rate": 9.30930930930931e-06,
      "loss": 1.5011,
      "step": 180
    },
    {
      "epoch": 3.02,
      "learning_rate": 9.20920920920921e-06,
      "loss": 1.4947,
      "step": 190
    },
    {
      "epoch": 3.03,
      "learning_rate": 9.10910910910911e-06,
      "loss": 1.4436,
      "step": 200
    },
    {
      "epoch": 3.04,
      "learning_rate": 9.00900900900901e-06,
      "loss": 1.3674,
      "step": 210
    },
    {
      "epoch": 3.05,
      "learning_rate": 8.90890890890891e-06,
      "loss": 1.4239,
      "step": 220
    },
    {
      "epoch": 3.05,
      "eval_accuracy": 0.6359447004608295,
      "eval_loss": 1.2076131105422974,
      "eval_runtime": 167.766,
      "eval_samples_per_second": 1.293,
      "eval_steps_per_second": 0.167,
      "step": 224
    },
    {
      "epoch": 4.01,
      "learning_rate": 8.80880880880881e-06,
      "loss": 1.4136,
      "step": 230
    },
    {
      "epoch": 4.01,
      "learning_rate": 8.70870870870871e-06,
      "loss": 1.3031,
      "step": 240
    },
    {
      "epoch": 4.02,
      "learning_rate": 8.60860860860861e-06,
      "loss": 1.3306,
      "step": 250
    },
    {
      "epoch": 4.03,
      "learning_rate": 8.50850850850851e-06,
      "loss": 1.2416,
      "step": 260
    },
    {
      "epoch": 4.04,
      "learning_rate": 8.408408408408409e-06,
      "loss": 1.1851,
      "step": 270
    },
    {
      "epoch": 4.05,
      "learning_rate": 8.308308308308309e-06,
      "loss": 1.1884,
      "step": 280
    },
    {
      "epoch": 4.05,
      "eval_accuracy": 0.5576036866359447,
      "eval_loss": 1.1865636110305786,
      "eval_runtime": 165.9475,
      "eval_samples_per_second": 1.308,
      "eval_steps_per_second": 0.169,
      "step": 280
    },
    {
      "epoch": 5.01,
      "learning_rate": 8.208208208208209e-06,
      "loss": 1.3263,
      "step": 290
    },
    {
      "epoch": 5.02,
      "learning_rate": 8.108108108108109e-06,
      "loss": 1.2252,
      "step": 300
    },
    {
      "epoch": 5.03,
      "learning_rate": 8.00800800800801e-06,
      "loss": 1.2029,
      "step": 310
    },
    {
      "epoch": 5.04,
      "learning_rate": 7.90790790790791e-06,
      "loss": 1.1463,
      "step": 320
    },
    {
      "epoch": 5.05,
      "learning_rate": 7.807807807807808e-06,
      "loss": 1.2336,
      "step": 330
    },
    {
      "epoch": 5.05,
      "eval_accuracy": 0.6405529953917051,
      "eval_loss": 1.028122067451477,
      "eval_runtime": 165.6809,
      "eval_samples_per_second": 1.31,
      "eval_steps_per_second": 0.169,
      "step": 336
    },
    {
      "epoch": 6.0,
      "learning_rate": 7.707707707707708e-06,
      "loss": 1.224,
      "step": 340
    },
    {
      "epoch": 6.01,
      "learning_rate": 7.607607607607608e-06,
      "loss": 1.1512,
      "step": 350
    },
    {
      "epoch": 6.02,
      "learning_rate": 7.507507507507507e-06,
      "loss": 1.1913,
      "step": 360
    },
    {
      "epoch": 6.03,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 1.0144,
      "step": 370
    },
    {
      "epoch": 6.04,
      "learning_rate": 7.307307307307308e-06,
      "loss": 1.0578,
      "step": 380
    },
    {
      "epoch": 6.05,
      "learning_rate": 7.207207207207208e-06,
      "loss": 1.0726,
      "step": 390
    },
    {
      "epoch": 6.05,
      "eval_accuracy": 0.663594470046083,
      "eval_loss": 0.9172731637954712,
      "eval_runtime": 166.3821,
      "eval_samples_per_second": 1.304,
      "eval_steps_per_second": 0.168,
      "step": 392
    },
    {
      "epoch": 7.01,
      "learning_rate": 7.107107107107107e-06,
      "loss": 1.084,
      "step": 400
    },
    {
      "epoch": 7.02,
      "learning_rate": 7.007007007007007e-06,
      "loss": 1.0031,
      "step": 410
    },
    {
      "epoch": 7.03,
      "learning_rate": 6.906906906906907e-06,
      "loss": 0.997,
      "step": 420
    },
    {
      "epoch": 7.03,
      "learning_rate": 6.8068068068068075e-06,
      "loss": 1.1918,
      "step": 430
    },
    {
      "epoch": 7.04,
      "learning_rate": 6.706706706706707e-06,
      "loss": 1.0966,
      "step": 440
    },
    {
      "epoch": 7.05,
      "eval_accuracy": 0.5990783410138248,
      "eval_loss": 1.0597409009933472,
      "eval_runtime": 164.5381,
      "eval_samples_per_second": 1.319,
      "eval_steps_per_second": 0.17,
      "step": 448
    },
    {
      "epoch": 8.0,
      "learning_rate": 6.606606606606607e-06,
      "loss": 1.1823,
      "step": 450
    },
    {
      "epoch": 8.01,
      "learning_rate": 6.506506506506507e-06,
      "loss": 0.9391,
      "step": 460
    },
    {
      "epoch": 8.02,
      "learning_rate": 6.406406406406407e-06,
      "loss": 1.1039,
      "step": 470
    },
    {
      "epoch": 8.03,
      "learning_rate": 6.3063063063063065e-06,
      "loss": 1.1439,
      "step": 480
    },
    {
      "epoch": 8.04,
      "learning_rate": 6.206206206206207e-06,
      "loss": 0.9368,
      "step": 490
    },
    {
      "epoch": 8.05,
      "learning_rate": 6.106106106106107e-06,
      "loss": 0.956,
      "step": 500
    },
    {
      "epoch": 8.05,
      "eval_accuracy": 0.6682027649769585,
      "eval_loss": 0.9277688264846802,
      "eval_runtime": 165.822,
      "eval_samples_per_second": 1.309,
      "eval_steps_per_second": 0.169,
      "step": 504
    },
    {
      "epoch": 9.01,
      "learning_rate": 6.006006006006007e-06,
      "loss": 0.9728,
      "step": 510
    },
    {
      "epoch": 9.01,
      "learning_rate": 5.905905905905906e-06,
      "loss": 1.1136,
      "step": 520
    },
    {
      "epoch": 9.02,
      "learning_rate": 5.805805805805806e-06,
      "loss": 0.9854,
      "step": 530
    },
    {
      "epoch": 9.03,
      "learning_rate": 5.7057057057057065e-06,
      "loss": 0.9559,
      "step": 540
    },
    {
      "epoch": 9.04,
      "learning_rate": 5.605605605605607e-06,
      "loss": 0.9154,
      "step": 550
    },
    {
      "epoch": 9.05,
      "learning_rate": 5.505505505505506e-06,
      "loss": 1.0813,
      "step": 560
    },
    {
      "epoch": 9.05,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 1.0024998188018799,
      "eval_runtime": 167.438,
      "eval_samples_per_second": 1.296,
      "eval_steps_per_second": 0.167,
      "step": 560
    },
    {
      "epoch": 10.01,
      "learning_rate": 5.405405405405406e-06,
      "loss": 0.949,
      "step": 570
    },
    {
      "epoch": 10.02,
      "learning_rate": 5.305305305305306e-06,
      "loss": 0.9489,
      "step": 580
    },
    {
      "epoch": 10.03,
      "learning_rate": 5.205205205205206e-06,
      "loss": 1.0595,
      "step": 590
    },
    {
      "epoch": 10.04,
      "learning_rate": 5.105105105105106e-06,
      "loss": 0.8802,
      "step": 600
    },
    {
      "epoch": 10.05,
      "learning_rate": 5.005005005005006e-06,
      "loss": 0.8996,
      "step": 610
    },
    {
      "epoch": 10.05,
      "eval_accuracy": 0.7050691244239631,
      "eval_loss": 0.788213849067688,
      "eval_runtime": 164.6073,
      "eval_samples_per_second": 1.318,
      "eval_steps_per_second": 0.17,
      "step": 616
    },
    {
      "epoch": 11.0,
      "learning_rate": 4.904904904904905e-06,
      "loss": 0.9075,
      "step": 620
    },
    {
      "epoch": 11.01,
      "learning_rate": 4.804804804804805e-06,
      "loss": 0.921,
      "step": 630
    },
    {
      "epoch": 11.02,
      "learning_rate": 4.704704704704705e-06,
      "loss": 0.93,
      "step": 640
    },
    {
      "epoch": 11.03,
      "learning_rate": 4.604604604604605e-06,
      "loss": 0.7948,
      "step": 650
    },
    {
      "epoch": 11.04,
      "learning_rate": 4.504504504504505e-06,
      "loss": 0.9116,
      "step": 660
    },
    {
      "epoch": 11.05,
      "learning_rate": 4.404404404404405e-06,
      "loss": 0.947,
      "step": 670
    },
    {
      "epoch": 11.05,
      "eval_accuracy": 0.7188940092165899,
      "eval_loss": 0.8638418316841125,
      "eval_runtime": 164.8412,
      "eval_samples_per_second": 1.316,
      "eval_steps_per_second": 0.17,
      "step": 672
    },
    {
      "epoch": 12.01,
      "learning_rate": 4.304304304304305e-06,
      "loss": 0.9901,
      "step": 680
    },
    {
      "epoch": 12.02,
      "learning_rate": 4.204204204204204e-06,
      "loss": 0.8267,
      "step": 690
    },
    {
      "epoch": 12.03,
      "learning_rate": 4.1041041041041045e-06,
      "loss": 0.9112,
      "step": 700
    },
    {
      "epoch": 12.03,
      "learning_rate": 4.004004004004005e-06,
      "loss": 0.8341,
      "step": 710
    },
    {
      "epoch": 12.04,
      "learning_rate": 3.903903903903904e-06,
      "loss": 0.9386,
      "step": 720
    },
    {
      "epoch": 12.05,
      "eval_accuracy": 0.7004608294930875,
      "eval_loss": 0.8191553354263306,
      "eval_runtime": 164.8134,
      "eval_samples_per_second": 1.317,
      "eval_steps_per_second": 0.17,
      "step": 728
    },
    {
      "epoch": 13.0,
      "learning_rate": 3.803803803803804e-06,
      "loss": 0.9771,
      "step": 730
    },
    {
      "epoch": 13.01,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.8544,
      "step": 740
    },
    {
      "epoch": 13.02,
      "learning_rate": 3.603603603603604e-06,
      "loss": 0.997,
      "step": 750
    },
    {
      "epoch": 13.03,
      "learning_rate": 3.5035035035035036e-06,
      "loss": 0.9852,
      "step": 760
    },
    {
      "epoch": 13.04,
      "learning_rate": 3.4034034034034037e-06,
      "loss": 0.946,
      "step": 770
    },
    {
      "epoch": 13.05,
      "learning_rate": 3.3033033033033035e-06,
      "loss": 0.8754,
      "step": 780
    },
    {
      "epoch": 13.05,
      "eval_accuracy": 0.695852534562212,
      "eval_loss": 0.7741014361381531,
      "eval_runtime": 164.9986,
      "eval_samples_per_second": 1.315,
      "eval_steps_per_second": 0.17,
      "step": 784
    },
    {
      "epoch": 14.01,
      "learning_rate": 3.2032032032032036e-06,
      "loss": 0.9146,
      "step": 790
    },
    {
      "epoch": 14.01,
      "learning_rate": 3.1031031031031033e-06,
      "loss": 0.8221,
      "step": 800
    },
    {
      "epoch": 14.02,
      "learning_rate": 3.0030030030030034e-06,
      "loss": 0.8436,
      "step": 810
    },
    {
      "epoch": 14.03,
      "learning_rate": 2.902902902902903e-06,
      "loss": 0.812,
      "step": 820
    },
    {
      "epoch": 14.04,
      "learning_rate": 2.8028028028028033e-06,
      "loss": 0.8285,
      "step": 830
    },
    {
      "epoch": 14.05,
      "learning_rate": 2.702702702702703e-06,
      "loss": 0.8028,
      "step": 840
    },
    {
      "epoch": 14.05,
      "eval_accuracy": 0.7096774193548387,
      "eval_loss": 0.748151421546936,
      "eval_runtime": 165.2783,
      "eval_samples_per_second": 1.313,
      "eval_steps_per_second": 0.169,
      "step": 840
    },
    {
      "epoch": 15.01,
      "learning_rate": 2.602602602602603e-06,
      "loss": 0.9496,
      "step": 850
    },
    {
      "epoch": 15.02,
      "learning_rate": 2.502502502502503e-06,
      "loss": 0.8661,
      "step": 860
    },
    {
      "epoch": 15.03,
      "learning_rate": 2.4024024024024026e-06,
      "loss": 0.8948,
      "step": 870
    },
    {
      "epoch": 15.04,
      "learning_rate": 2.3023023023023023e-06,
      "loss": 0.7513,
      "step": 880
    },
    {
      "epoch": 15.05,
      "learning_rate": 2.2022022022022024e-06,
      "loss": 0.7038,
      "step": 890
    },
    {
      "epoch": 15.05,
      "eval_accuracy": 0.7327188940092166,
      "eval_loss": 0.7463698983192444,
      "eval_runtime": 165.2407,
      "eval_samples_per_second": 1.313,
      "eval_steps_per_second": 0.169,
      "step": 896
    },
    {
      "epoch": 16.0,
      "learning_rate": 2.102102102102102e-06,
      "loss": 0.8482,
      "step": 900
    },
    {
      "epoch": 16.01,
      "learning_rate": 2.0020020020020023e-06,
      "loss": 0.7576,
      "step": 910
    },
    {
      "epoch": 16.02,
      "learning_rate": 1.901901901901902e-06,
      "loss": 0.8506,
      "step": 920
    },
    {
      "epoch": 16.03,
      "learning_rate": 1.801801801801802e-06,
      "loss": 0.8435,
      "step": 930
    },
    {
      "epoch": 16.04,
      "learning_rate": 1.7017017017017019e-06,
      "loss": 0.9429,
      "step": 940
    },
    {
      "epoch": 16.05,
      "learning_rate": 1.6016016016016018e-06,
      "loss": 0.6302,
      "step": 950
    },
    {
      "epoch": 16.05,
      "eval_accuracy": 0.7004608294930875,
      "eval_loss": 0.7732542753219604,
      "eval_runtime": 166.0375,
      "eval_samples_per_second": 1.307,
      "eval_steps_per_second": 0.169,
      "step": 952
    },
    {
      "epoch": 17.01,
      "learning_rate": 1.5015015015015017e-06,
      "loss": 0.8215,
      "step": 960
    },
    {
      "epoch": 17.02,
      "learning_rate": 1.4014014014014016e-06,
      "loss": 0.7503,
      "step": 970
    },
    {
      "epoch": 17.03,
      "learning_rate": 1.3013013013013016e-06,
      "loss": 0.8131,
      "step": 980
    },
    {
      "epoch": 17.03,
      "learning_rate": 1.2012012012012013e-06,
      "loss": 0.7954,
      "step": 990
    },
    {
      "epoch": 17.04,
      "learning_rate": 1.1011011011011012e-06,
      "loss": 0.8387,
      "step": 1000
    },
    {
      "epoch": 17.05,
      "eval_accuracy": 0.7235023041474654,
      "eval_loss": 0.722377598285675,
      "eval_runtime": 164.6978,
      "eval_samples_per_second": 1.318,
      "eval_steps_per_second": 0.17,
      "step": 1008
    },
    {
      "epoch": 18.0,
      "learning_rate": 1.0010010010010011e-06,
      "loss": 0.9197,
      "step": 1010
    },
    {
      "epoch": 18.01,
      "learning_rate": 9.00900900900901e-07,
      "loss": 0.8602,
      "step": 1020
    },
    {
      "epoch": 18.02,
      "learning_rate": 8.008008008008009e-07,
      "loss": 0.706,
      "step": 1030
    },
    {
      "epoch": 18.03,
      "learning_rate": 7.007007007007008e-07,
      "loss": 0.878,
      "step": 1040
    },
    {
      "epoch": 18.04,
      "learning_rate": 6.006006006006006e-07,
      "loss": 0.8754,
      "step": 1050
    },
    {
      "epoch": 18.05,
      "learning_rate": 5.005005005005006e-07,
      "loss": 0.5853,
      "step": 1060
    },
    {
      "epoch": 18.05,
      "eval_accuracy": 0.7142857142857143,
      "eval_loss": 0.7359205484390259,
      "eval_runtime": 165.3361,
      "eval_samples_per_second": 1.312,
      "eval_steps_per_second": 0.169,
      "step": 1064
    },
    {
      "epoch": 19.01,
      "learning_rate": 4.0040040040040045e-07,
      "loss": 0.7846,
      "step": 1070
    },
    {
      "epoch": 19.01,
      "learning_rate": 3.003003003003003e-07,
      "loss": 0.9235,
      "step": 1080
    },
    {
      "epoch": 19.02,
      "learning_rate": 2.0020020020020022e-07,
      "loss": 0.9745,
      "step": 1090
    },
    {
      "epoch": 19.03,
      "learning_rate": 1.0010010010010011e-07,
      "loss": 0.6344,
      "step": 1100
    },
    {
      "epoch": 19.04,
      "learning_rate": 0.0,
      "loss": 0.7482,
      "step": 1110
    },
    {
      "epoch": 19.04,
      "eval_accuracy": 0.7188940092165899,
      "eval_loss": 0.7369951605796814,
      "eval_runtime": 166.4385,
      "eval_samples_per_second": 1.304,
      "eval_steps_per_second": 0.168,
      "step": 1110
    },
    {
      "epoch": 19.04,
      "step": 1110,
      "total_flos": 1.1041673732998595e+19,
      "train_loss": 1.081317645150262,
      "train_runtime": 11661.5957,
      "train_samples_per_second": 0.761,
      "train_steps_per_second": 0.095
    },
    {
      "epoch": 19.04,
      "eval_accuracy": 0.7129629629629629,
      "eval_loss": 0.7363542914390564,
      "eval_runtime": 167.4287,
      "eval_samples_per_second": 1.29,
      "eval_steps_per_second": 0.161,
      "step": 1110
    },
    {
      "epoch": 19.04,
      "eval_accuracy": 0.7129629629629629,
      "eval_loss": 0.7363542914390564,
      "eval_runtime": 168.5999,
      "eval_samples_per_second": 1.281,
      "eval_steps_per_second": 0.16,
      "step": 1110
    }
  ],
  "logging_steps": 10,
  "max_steps": 1110,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 1.1041673732998595e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}