{
  "best_metric": 0.9504132231404959,
  "best_model_checkpoint": "videomae-base-finetuned-good-gestureUnitsV3/checkpoint-255",
  "epoch": 19.05,
  "eval_steps": 500,
  "global_step": 340,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.029411764705882353,
      "grad_norm": 9.534835815429688,
      "learning_rate": 1.4705882352941177e-06,
      "loss": 0.7047,
      "step": 10
    },
    {
      "epoch": 0.05,
      "eval_accuracy": 0.5619834710743802,
      "eval_accuracy_gunit": 0.575,
      "eval_accuracy_nothing": 0.5555555555555556,
      "eval_loss": 0.7023928165435791,
      "eval_runtime": 15.7256,
      "eval_samples_per_second": 7.694,
      "eval_steps_per_second": 1.017,
      "step": 17
    },
    {
      "epoch": 1.0088235294117647,
      "grad_norm": 5.451850414276123,
      "learning_rate": 2.9411764705882355e-06,
      "loss": 0.7083,
      "step": 20
    },
    {
      "epoch": 1.0382352941176471,
      "grad_norm": 5.070091724395752,
      "learning_rate": 4.411764705882353e-06,
      "loss": 0.6958,
      "step": 30
    },
    {
      "epoch": 1.05,
      "eval_accuracy": 0.38016528925619836,
      "eval_accuracy_gunit": 0.725,
      "eval_accuracy_nothing": 0.20987654320987653,
      "eval_loss": 0.7165102362632751,
      "eval_runtime": 15.8798,
      "eval_samples_per_second": 7.62,
      "eval_steps_per_second": 1.008,
      "step": 34
    },
    {
      "epoch": 2.0176470588235293,
      "grad_norm": 4.038697719573975,
      "learning_rate": 4.901960784313726e-06,
      "loss": 0.6585,
      "step": 40
    },
    {
      "epoch": 2.0470588235294116,
      "grad_norm": 3.2235519886016846,
      "learning_rate": 4.7385620915032685e-06,
      "loss": 0.6534,
      "step": 50
    },
    {
      "epoch": 2.05,
      "eval_accuracy": 0.371900826446281,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 0.13580246913580246,
      "eval_loss": 0.7150475382804871,
      "eval_runtime": 16.0184,
      "eval_samples_per_second": 7.554,
      "eval_steps_per_second": 0.999,
      "step": 51
    },
    {
      "epoch": 3.026470588235294,
      "grad_norm": 4.749432563781738,
      "learning_rate": 4.5751633986928105e-06,
      "loss": 0.655,
      "step": 60
    },
    {
      "epoch": 3.05,
      "eval_accuracy": 0.32231404958677684,
      "eval_accuracy_gunit": 0.975,
      "eval_accuracy_nothing": 0.0,
      "eval_loss": 0.7505635619163513,
      "eval_runtime": 15.7625,
      "eval_samples_per_second": 7.676,
      "eval_steps_per_second": 1.015,
      "step": 68
    },
    {
      "epoch": 4.0058823529411764,
      "grad_norm": 6.722472190856934,
      "learning_rate": 4.411764705882353e-06,
      "loss": 0.6741,
      "step": 70
    },
    {
      "epoch": 4.035294117647059,
      "grad_norm": 6.8286638259887695,
      "learning_rate": 4.2483660130718954e-06,
      "loss": 0.6545,
      "step": 80
    },
    {
      "epoch": 4.05,
      "eval_accuracy": 0.4297520661157025,
      "eval_accuracy_gunit": 0.95,
      "eval_accuracy_nothing": 0.1728395061728395,
      "eval_loss": 0.6991418600082397,
      "eval_runtime": 15.5171,
      "eval_samples_per_second": 7.798,
      "eval_steps_per_second": 1.031,
      "step": 85
    },
    {
      "epoch": 5.014705882352941,
      "grad_norm": 7.191628456115723,
      "learning_rate": 4.084967320261438e-06,
      "loss": 0.56,
      "step": 90
    },
    {
      "epoch": 5.044117647058823,
      "grad_norm": 5.416815757751465,
      "learning_rate": 3.92156862745098e-06,
      "loss": 0.6321,
      "step": 100
    },
    {
      "epoch": 5.05,
      "eval_accuracy": 0.47107438016528924,
      "eval_accuracy_gunit": 0.95,
      "eval_accuracy_nothing": 0.2345679012345679,
      "eval_loss": 0.6750265955924988,
      "eval_runtime": 15.9488,
      "eval_samples_per_second": 7.587,
      "eval_steps_per_second": 1.003,
      "step": 102
    },
    {
      "epoch": 6.023529411764706,
      "grad_norm": 4.386753559112549,
      "learning_rate": 3.758169934640523e-06,
      "loss": 0.5883,
      "step": 110
    },
    {
      "epoch": 6.05,
      "eval_accuracy": 0.743801652892562,
      "eval_accuracy_gunit": 0.9,
      "eval_accuracy_nothing": 0.6666666666666666,
      "eval_loss": 0.6142133474349976,
      "eval_runtime": 16.6481,
      "eval_samples_per_second": 7.268,
      "eval_steps_per_second": 0.961,
      "step": 119
    },
    {
      "epoch": 7.002941176470588,
      "grad_norm": 5.5603108406066895,
      "learning_rate": 3.5947712418300657e-06,
      "loss": 0.5803,
      "step": 120
    },
    {
      "epoch": 7.0323529411764705,
      "grad_norm": 3.9502112865448,
      "learning_rate": 3.431372549019608e-06,
      "loss": 0.5045,
      "step": 130
    },
    {
      "epoch": 7.05,
      "eval_accuracy": 0.8181818181818182,
      "eval_accuracy_gunit": 0.9,
      "eval_accuracy_nothing": 0.7777777777777778,
      "eval_loss": 0.549534022808075,
      "eval_runtime": 15.8267,
      "eval_samples_per_second": 7.645,
      "eval_steps_per_second": 1.011,
      "step": 136
    },
    {
      "epoch": 8.011764705882353,
      "grad_norm": 5.036988735198975,
      "learning_rate": 3.2679738562091506e-06,
      "loss": 0.578,
      "step": 140
    },
    {
      "epoch": 8.041176470588235,
      "grad_norm": 4.750277519226074,
      "learning_rate": 3.104575163398693e-06,
      "loss": 0.455,
      "step": 150
    },
    {
      "epoch": 8.05,
      "eval_accuracy": 0.71900826446281,
      "eval_accuracy_gunit": 0.925,
      "eval_accuracy_nothing": 0.6172839506172839,
      "eval_loss": 0.572287917137146,
      "eval_runtime": 16.0277,
      "eval_samples_per_second": 7.549,
      "eval_steps_per_second": 0.998,
      "step": 153
    },
    {
      "epoch": 9.020588235294118,
      "grad_norm": 4.390650749206543,
      "learning_rate": 2.9411764705882355e-06,
      "loss": 0.382,
      "step": 160
    },
    {
      "epoch": 9.05,
      "grad_norm": 12.403214454650879,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.4191,
      "step": 170
    },
    {
      "epoch": 9.05,
      "eval_accuracy": 0.9338842975206612,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 0.9753086419753086,
      "eval_loss": 0.37085020542144775,
      "eval_runtime": 19.423,
      "eval_samples_per_second": 6.23,
      "eval_steps_per_second": 0.824,
      "step": 170
    },
    {
      "epoch": 10.029411764705882,
      "grad_norm": 8.752729415893555,
      "learning_rate": 2.6143790849673208e-06,
      "loss": 0.365,
      "step": 180
    },
    {
      "epoch": 10.05,
      "eval_accuracy": 0.9090909090909091,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 0.9382716049382716,
      "eval_loss": 0.3546590209007263,
      "eval_runtime": 15.5146,
      "eval_samples_per_second": 7.799,
      "eval_steps_per_second": 1.031,
      "step": 187
    },
    {
      "epoch": 11.008823529411766,
      "grad_norm": 6.302486896514893,
      "learning_rate": 2.450980392156863e-06,
      "loss": 0.3493,
      "step": 190
    },
    {
      "epoch": 11.038235294117648,
      "grad_norm": 4.457513809204102,
      "learning_rate": 2.2875816993464053e-06,
      "loss": 0.2593,
      "step": 200
    },
    {
      "epoch": 11.05,
      "eval_accuracy": 0.8925619834710744,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 0.9135802469135802,
      "eval_loss": 0.36317938566207886,
      "eval_runtime": 15.7061,
      "eval_samples_per_second": 7.704,
      "eval_steps_per_second": 1.019,
      "step": 204
    },
    {
      "epoch": 12.01764705882353,
      "grad_norm": 11.817601203918457,
      "learning_rate": 2.1241830065359477e-06,
      "loss": 0.3111,
      "step": 210
    },
    {
      "epoch": 12.047058823529412,
      "grad_norm": 6.769512176513672,
      "learning_rate": 1.96078431372549e-06,
      "loss": 0.2225,
      "step": 220
    },
    {
      "epoch": 12.05,
      "eval_accuracy": 0.9421487603305785,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 0.9876543209876543,
      "eval_loss": 0.24565516412258148,
      "eval_runtime": 20.1411,
      "eval_samples_per_second": 6.008,
      "eval_steps_per_second": 0.794,
      "step": 221
    },
    {
      "epoch": 13.026470588235295,
      "grad_norm": 3.0028862953186035,
      "learning_rate": 1.7973856209150328e-06,
      "loss": 0.2121,
      "step": 230
    },
    {
      "epoch": 13.05,
      "eval_accuracy": 0.9256198347107438,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 0.9629629629629629,
      "eval_loss": 0.26187241077423096,
      "eval_runtime": 15.6231,
      "eval_samples_per_second": 7.745,
      "eval_steps_per_second": 1.024,
      "step": 238
    },
    {
      "epoch": 14.005882352941176,
      "grad_norm": 11.049546241760254,
      "learning_rate": 1.6339869281045753e-06,
      "loss": 0.1814,
      "step": 240
    },
    {
      "epoch": 14.035294117647059,
      "grad_norm": 13.598651885986328,
      "learning_rate": 1.4705882352941177e-06,
      "loss": 0.1506,
      "step": 250
    },
    {
      "epoch": 14.05,
      "eval_accuracy": 0.9504132231404959,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 1.0,
      "eval_loss": 0.2190253734588623,
      "eval_runtime": 15.7623,
      "eval_samples_per_second": 7.677,
      "eval_steps_per_second": 1.015,
      "step": 255
    },
    {
      "epoch": 15.014705882352942,
      "grad_norm": 5.206620693206787,
      "learning_rate": 1.3071895424836604e-06,
      "loss": 0.1518,
      "step": 260
    },
    {
      "epoch": 15.044117647058824,
      "grad_norm": 15.521167755126953,
      "learning_rate": 1.1437908496732026e-06,
      "loss": 0.1335,
      "step": 270
    },
    {
      "epoch": 15.05,
      "eval_accuracy": 0.9338842975206612,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 0.9753086419753086,
      "eval_loss": 0.22213229537010193,
      "eval_runtime": 15.6243,
      "eval_samples_per_second": 7.744,
      "eval_steps_per_second": 1.024,
      "step": 272
    },
    {
      "epoch": 16.023529411764706,
      "grad_norm": 3.6706037521362305,
      "learning_rate": 9.80392156862745e-07,
      "loss": 0.1039,
      "step": 280
    },
    {
      "epoch": 16.05,
      "eval_accuracy": 0.9421487603305785,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 0.9876543209876543,
      "eval_loss": 0.2107107788324356,
      "eval_runtime": 15.8386,
      "eval_samples_per_second": 7.64,
      "eval_steps_per_second": 1.01,
      "step": 289
    },
    {
      "epoch": 17.00294117647059,
      "grad_norm": 2.560628890991211,
      "learning_rate": 8.169934640522876e-07,
      "loss": 0.1959,
      "step": 290
    },
    {
      "epoch": 17.03235294117647,
      "grad_norm": 13.137368202209473,
      "learning_rate": 6.535947712418302e-07,
      "loss": 0.1036,
      "step": 300
    },
    {
      "epoch": 17.05,
      "eval_accuracy": 0.9421487603305785,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 0.9876543209876543,
      "eval_loss": 0.20594191551208496,
      "eval_runtime": 15.7489,
      "eval_samples_per_second": 7.683,
      "eval_steps_per_second": 1.016,
      "step": 306
    },
    {
      "epoch": 18.011764705882353,
      "grad_norm": 105.94062805175781,
      "learning_rate": 4.901960784313725e-07,
      "loss": 0.139,
      "step": 310
    },
    {
      "epoch": 18.041176470588237,
      "grad_norm": 7.893646240234375,
      "learning_rate": 3.267973856209151e-07,
      "loss": 0.1019,
      "step": 320
    },
    {
      "epoch": 18.05,
      "eval_accuracy": 0.9421487603305785,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 0.9876543209876543,
      "eval_loss": 0.2062729150056839,
      "eval_runtime": 15.7147,
      "eval_samples_per_second": 7.7,
      "eval_steps_per_second": 1.018,
      "step": 323
    },
    {
      "epoch": 19.020588235294117,
      "grad_norm": 4.479135036468506,
      "learning_rate": 1.6339869281045755e-07,
      "loss": 0.0837,
      "step": 330
    },
    {
      "epoch": 19.05,
      "grad_norm": 84.62340545654297,
      "learning_rate": 0.0,
      "loss": 0.1181,
      "step": 340
    },
    {
      "epoch": 19.05,
      "eval_accuracy": 0.9504132231404959,
      "eval_accuracy_gunit": 0.85,
      "eval_accuracy_nothing": 1.0,
      "eval_loss": 0.21026436984539032,
      "eval_runtime": 15.8645,
      "eval_samples_per_second": 7.627,
      "eval_steps_per_second": 1.009,
      "step": 340
    },
    {
      "epoch": 19.05,
      "step": 340,
      "total_flos": 3.339453099784274e+18,
      "train_loss": 0.3907703813384561,
      "train_runtime": 996.3737,
      "train_samples_per_second": 2.73,
      "train_steps_per_second": 0.341
    },
    {
      "epoch": 19.05,
      "eval_accuracy": 0.8932038834951457,
      "eval_accuracy_gunit": 0.5416666666666666,
      "eval_accuracy_nothing": 1.0,
      "eval_loss": 0.34902578592300415,
      "eval_runtime": 14.5221,
      "eval_samples_per_second": 7.093,
      "eval_steps_per_second": 0.895,
      "step": 340
    },
    {
      "epoch": 19.05,
      "eval_accuracy": 0.8932038834951457,
      "eval_accuracy_gunit": 0.5416666666666666,
      "eval_accuracy_nothing": 1.0,
      "eval_loss": 0.3490257263183594,
      "eval_runtime": 13.7098,
      "eval_samples_per_second": 7.513,
      "eval_steps_per_second": 0.948,
      "step": 340
    }
  ],
  "logging_steps": 10,
  "max_steps": 340,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.339453099784274e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}