{
  "best_metric": 0.8297674418604651,
  "best_model_checkpoint": "video/train/checkpoint/videomae-base/ucf101_42/checkpoint-5700",
  "epoch": 19.871751886001675,
  "eval_steps": 500,
  "global_step": 5960,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.01,
      "eval_accuracy": 0.4886511627906977,
      "eval_loss": 2.334453582763672,
      "eval_runtime": 390.4063,
      "eval_samples_per_second": 9.69,
      "eval_steps_per_second": 0.154,
      "step": 300
    },
    {
      "epoch": 2.01,
      "eval_accuracy": 0.6433488372093024,
      "eval_loss": 1.4956246614456177,
      "eval_runtime": 378.3625,
      "eval_samples_per_second": 9.998,
      "eval_steps_per_second": 0.159,
      "step": 600
    },
    {
      "epoch": 3.01,
      "eval_accuracy": 0.7161860465116279,
      "eval_loss": 1.1646597385406494,
      "eval_runtime": 374.5393,
      "eval_samples_per_second": 10.1,
      "eval_steps_per_second": 0.16,
      "step": 900
    },
    {
      "epoch": 3.34,
      "grad_norm": 6.9290690422058105,
      "learning_rate": 4.161073825503356e-05,
      "loss": 1.5784,
      "step": 1000
    },
    {
      "epoch": 4.01,
      "eval_accuracy": 0.7028837209302325,
      "eval_loss": 1.1531013250350952,
      "eval_runtime": 376.3709,
      "eval_samples_per_second": 10.051,
      "eval_steps_per_second": 0.159,
      "step": 1200
    },
    {
      "epoch": 5.01,
      "eval_accuracy": 0.7333953488372092,
      "eval_loss": 1.0251258611679077,
      "eval_runtime": 377.2789,
      "eval_samples_per_second": 10.027,
      "eval_steps_per_second": 0.159,
      "step": 1500
    },
    {
      "epoch": 6.01,
      "eval_accuracy": 0.7333023255813953,
      "eval_loss": 1.031457781791687,
      "eval_runtime": 376.6891,
      "eval_samples_per_second": 10.043,
      "eval_steps_per_second": 0.159,
      "step": 1800
    },
    {
      "epoch": 6.67,
      "grad_norm": 7.911869049072266,
      "learning_rate": 3.3221476510067115e-05,
      "loss": 0.1821,
      "step": 2000
    },
    {
      "epoch": 7.01,
      "eval_accuracy": 0.7616744186046511,
      "eval_loss": 0.9787126779556274,
      "eval_runtime": 379.965,
      "eval_samples_per_second": 9.956,
      "eval_steps_per_second": 0.158,
      "step": 2100
    },
    {
      "epoch": 8.01,
      "eval_accuracy": 0.7838139534883721,
      "eval_loss": 0.8932655453681946,
      "eval_runtime": 375.8521,
      "eval_samples_per_second": 10.065,
      "eval_steps_per_second": 0.16,
      "step": 2400
    },
    {
      "epoch": 9.01,
      "eval_accuracy": 0.7917209302325582,
      "eval_loss": 0.8780561089515686,
      "eval_runtime": 373.4872,
      "eval_samples_per_second": 10.129,
      "eval_steps_per_second": 0.161,
      "step": 2700
    },
    {
      "epoch": 10.01,
      "grad_norm": 0.12537457048892975,
      "learning_rate": 2.4832214765100674e-05,
      "loss": 0.0651,
      "step": 3000
    },
    {
      "epoch": 10.01,
      "eval_accuracy": 0.7909767441860465,
      "eval_loss": 0.9051322937011719,
      "eval_runtime": 375.6037,
      "eval_samples_per_second": 10.072,
      "eval_steps_per_second": 0.16,
      "step": 3000
    },
    {
      "epoch": 11.01,
      "eval_accuracy": 0.790046511627907,
      "eval_loss": 0.9593069553375244,
      "eval_runtime": 375.7576,
      "eval_samples_per_second": 10.068,
      "eval_steps_per_second": 0.16,
      "step": 3300
    },
    {
      "epoch": 12.01,
      "eval_accuracy": 0.8186976744186046,
      "eval_loss": 0.8053779602050781,
      "eval_runtime": 377.8682,
      "eval_samples_per_second": 10.011,
      "eval_steps_per_second": 0.159,
      "step": 3600
    },
    {
      "epoch": 13.01,
      "eval_accuracy": 0.8142325581395349,
      "eval_loss": 0.8678916692733765,
      "eval_runtime": 376.9615,
      "eval_samples_per_second": 10.036,
      "eval_steps_per_second": 0.159,
      "step": 3900
    },
    {
      "epoch": 13.34,
      "grad_norm": 10.020633697509766,
      "learning_rate": 1.644295302013423e-05,
      "loss": 0.0265,
      "step": 4000
    },
    {
      "epoch": 14.01,
      "eval_accuracy": 0.8208372093023256,
      "eval_loss": 0.8379742503166199,
      "eval_runtime": 373.9706,
      "eval_samples_per_second": 10.116,
      "eval_steps_per_second": 0.16,
      "step": 4200
    },
    {
      "epoch": 15.01,
      "eval_accuracy": 0.8246511627906977,
      "eval_loss": 0.8317446708679199,
      "eval_runtime": 375.9164,
      "eval_samples_per_second": 10.063,
      "eval_steps_per_second": 0.16,
      "step": 4500
    },
    {
      "epoch": 16.01,
      "eval_accuracy": 0.8249302325581396,
      "eval_loss": 0.8027446269989014,
      "eval_runtime": 375.7325,
      "eval_samples_per_second": 10.068,
      "eval_steps_per_second": 0.16,
      "step": 4800
    },
    {
      "epoch": 16.67,
      "grad_norm": 0.026239760220050812,
      "learning_rate": 8.053691275167785e-06,
      "loss": 0.0091,
      "step": 5000
    },
    {
      "epoch": 17.01,
      "eval_accuracy": 0.8254883720930233,
      "eval_loss": 0.8239758610725403,
      "eval_runtime": 375.8655,
      "eval_samples_per_second": 10.065,
      "eval_steps_per_second": 0.16,
      "step": 5100
    },
    {
      "epoch": 18.01,
      "eval_accuracy": 0.8211162790697675,
      "eval_loss": 0.8480401039123535,
      "eval_runtime": 375.7779,
      "eval_samples_per_second": 10.067,
      "eval_steps_per_second": 0.16,
      "step": 5400
    },
    {
      "epoch": 19.01,
      "eval_accuracy": 0.8297674418604651,
      "eval_loss": 0.8198431134223938,
      "eval_runtime": 372.9411,
      "eval_samples_per_second": 10.144,
      "eval_steps_per_second": 0.161,
      "step": 5700
    },
    {
      "epoch": 19.87,
      "eval_accuracy": 0.8282790697674418,
      "eval_loss": 0.8315407037734985,
      "eval_runtime": 376.729,
      "eval_samples_per_second": 10.042,
      "eval_steps_per_second": 0.159,
      "step": 5960
    },
    {
      "epoch": 19.87,
      "step": 5960,
      "total_flos": 2.3636640203347614e+20,
      "train_loss": 0.3129509675422771,
      "train_runtime": 26481.3321,
      "train_samples_per_second": 7.203,
      "train_steps_per_second": 0.225
    }
  ],
  "logging_steps": 1000,
  "max_steps": 5960,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "total_flos": 2.3636640203347614e+20,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}