{
  "best_metric": 1.0,
  "best_model_checkpoint": "MCG-NJU/videomae-large/checkpoint-130",
  "epoch": 29.033333333333335,
  "eval_steps": 500,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 2.3203,
      "step": 10
    },
    {
      "epoch": 0.03,
      "eval_accuracy": 0.15714285714285714,
      "eval_loss": 2.199398994445801,
      "eval_runtime": 5.6646,
      "eval_samples_per_second": 12.357,
      "eval_steps_per_second": 0.53,
      "step": 10
    },
    {
      "epoch": 1.03,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.9795,
      "step": 20
    },
    {
      "epoch": 1.03,
      "eval_accuracy": 0.34285714285714286,
      "eval_loss": 1.7835056781768799,
      "eval_runtime": 6.2612,
      "eval_samples_per_second": 11.18,
      "eval_steps_per_second": 0.479,
      "step": 20
    },
    {
      "epoch": 2.03,
      "learning_rate": 5e-05,
      "loss": 1.0467,
      "step": 30
    },
    {
      "epoch": 2.03,
      "eval_accuracy": 0.6571428571428571,
      "eval_loss": 0.7310971617698669,
      "eval_runtime": 6.125,
      "eval_samples_per_second": 11.429,
      "eval_steps_per_second": 0.49,
      "step": 30
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.814814814814815e-05,
      "loss": 0.301,
      "step": 40
    },
    {
      "epoch": 3.03,
      "eval_accuracy": 0.9428571428571428,
      "eval_loss": 0.21951383352279663,
      "eval_runtime": 5.975,
      "eval_samples_per_second": 11.715,
      "eval_steps_per_second": 0.502,
      "step": 40
    },
    {
      "epoch": 4.03,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.1061,
      "step": 50
    },
    {
      "epoch": 4.03,
      "eval_accuracy": 0.9142857142857143,
      "eval_loss": 0.15290319919586182,
      "eval_runtime": 6.6932,
      "eval_samples_per_second": 10.458,
      "eval_steps_per_second": 0.448,
      "step": 50
    },
    {
      "epoch": 5.03,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.0499,
      "step": 60
    },
    {
      "epoch": 5.03,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.08257948607206345,
      "eval_runtime": 6.8638,
      "eval_samples_per_second": 10.198,
      "eval_steps_per_second": 0.437,
      "step": 60
    },
    {
      "epoch": 6.03,
      "learning_rate": 4.259259259259259e-05,
      "loss": 0.079,
      "step": 70
    },
    {
      "epoch": 6.03,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.05339507758617401,
      "eval_runtime": 5.6428,
      "eval_samples_per_second": 12.405,
      "eval_steps_per_second": 0.532,
      "step": 70
    },
    {
      "epoch": 7.03,
      "learning_rate": 4.074074074074074e-05,
      "loss": 0.0487,
      "step": 80
    },
    {
      "epoch": 7.03,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.02986798621714115,
      "eval_runtime": 5.2266,
      "eval_samples_per_second": 13.393,
      "eval_steps_per_second": 0.574,
      "step": 80
    },
    {
      "epoch": 8.03,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.0217,
      "step": 90
    },
    {
      "epoch": 8.03,
      "eval_accuracy": 0.9,
      "eval_loss": 0.32826170325279236,
      "eval_runtime": 6.6113,
      "eval_samples_per_second": 10.588,
      "eval_steps_per_second": 0.454,
      "step": 90
    },
    {
      "epoch": 9.03,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.0387,
      "step": 100
    },
    {
      "epoch": 9.03,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.026779260486364365,
      "eval_runtime": 6.3515,
      "eval_samples_per_second": 11.021,
      "eval_steps_per_second": 0.472,
      "step": 100
    },
    {
      "epoch": 10.03,
      "learning_rate": 3.518518518518519e-05,
      "loss": 0.0252,
      "step": 110
    },
    {
      "epoch": 10.03,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.03859327733516693,
      "eval_runtime": 6.9316,
      "eval_samples_per_second": 10.099,
      "eval_steps_per_second": 0.433,
      "step": 110
    },
    {
      "epoch": 11.03,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0324,
      "step": 120
    },
    {
      "epoch": 11.03,
      "eval_accuracy": 0.9,
      "eval_loss": 0.3067415654659271,
      "eval_runtime": 5.7803,
      "eval_samples_per_second": 12.11,
      "eval_steps_per_second": 0.519,
      "step": 120
    },
    {
      "epoch": 12.03,
      "learning_rate": 3.148148148148148e-05,
      "loss": 0.0022,
      "step": 130
    },
    {
      "epoch": 12.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.013092391192913055,
      "eval_runtime": 5.2039,
      "eval_samples_per_second": 13.451,
      "eval_steps_per_second": 0.576,
      "step": 130
    },
    {
      "epoch": 13.03,
      "learning_rate": 2.962962962962963e-05,
      "loss": 0.0115,
      "step": 140
    },
    {
      "epoch": 13.03,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.08892710506916046,
      "eval_runtime": 4.8696,
      "eval_samples_per_second": 14.375,
      "eval_steps_per_second": 0.616,
      "step": 140
    },
    {
      "epoch": 14.03,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.0225,
      "step": 150
    },
    {
      "epoch": 14.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.009070915170013905,
      "eval_runtime": 6.0016,
      "eval_samples_per_second": 11.663,
      "eval_steps_per_second": 0.5,
      "step": 150
    },
    {
      "epoch": 15.03,
      "learning_rate": 2.5925925925925925e-05,
      "loss": 0.0012,
      "step": 160
    },
    {
      "epoch": 15.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.008068457245826721,
      "eval_runtime": 5.6404,
      "eval_samples_per_second": 12.411,
      "eval_steps_per_second": 0.532,
      "step": 160
    },
    {
      "epoch": 16.03,
      "learning_rate": 2.4074074074074074e-05,
      "loss": 0.001,
      "step": 170
    },
    {
      "epoch": 16.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.010284548625349998,
      "eval_runtime": 5.7893,
      "eval_samples_per_second": 12.091,
      "eval_steps_per_second": 0.518,
      "step": 170
    },
    {
      "epoch": 17.03,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.0255,
      "step": 180
    },
    {
      "epoch": 17.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.01131558045744896,
      "eval_runtime": 6.4736,
      "eval_samples_per_second": 10.813,
      "eval_steps_per_second": 0.463,
      "step": 180
    },
    {
      "epoch": 18.03,
      "learning_rate": 2.037037037037037e-05,
      "loss": 0.0016,
      "step": 190
    },
    {
      "epoch": 18.03,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.025160381570458412,
      "eval_runtime": 5.7118,
      "eval_samples_per_second": 12.255,
      "eval_steps_per_second": 0.525,
      "step": 190
    },
    {
      "epoch": 19.03,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.0039,
      "step": 200
    },
    {
      "epoch": 19.03,
      "eval_accuracy": 0.9857142857142858,
      "eval_loss": 0.017666514962911606,
      "eval_runtime": 5.7688,
      "eval_samples_per_second": 12.134,
      "eval_steps_per_second": 0.52,
      "step": 200
    },
    {
      "epoch": 20.03,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.0007,
      "step": 210
    },
    {
      "epoch": 20.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0016855127178132534,
      "eval_runtime": 5.9439,
      "eval_samples_per_second": 11.777,
      "eval_steps_per_second": 0.505,
      "step": 210
    },
    {
      "epoch": 21.03,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 0.0006,
      "step": 220
    },
    {
      "epoch": 21.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.00132262974511832,
      "eval_runtime": 6.2152,
      "eval_samples_per_second": 11.263,
      "eval_steps_per_second": 0.483,
      "step": 220
    },
    {
      "epoch": 22.03,
      "learning_rate": 1.2962962962962962e-05,
      "loss": 0.0006,
      "step": 230
    },
    {
      "epoch": 22.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0012219419004395604,
      "eval_runtime": 5.788,
      "eval_samples_per_second": 12.094,
      "eval_steps_per_second": 0.518,
      "step": 230
    },
    {
      "epoch": 23.03,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.0007,
      "step": 240
    },
    {
      "epoch": 23.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.001058029243722558,
      "eval_runtime": 6.1445,
      "eval_samples_per_second": 11.392,
      "eval_steps_per_second": 0.488,
      "step": 240
    },
    {
      "epoch": 24.03,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.0005,
      "step": 250
    },
    {
      "epoch": 24.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0010857696179300547,
      "eval_runtime": 6.974,
      "eval_samples_per_second": 10.037,
      "eval_steps_per_second": 0.43,
      "step": 250
    },
    {
      "epoch": 25.03,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.0005,
      "step": 260
    },
    {
      "epoch": 25.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0010956026380881667,
      "eval_runtime": 5.6397,
      "eval_samples_per_second": 12.412,
      "eval_steps_per_second": 0.532,
      "step": 260
    },
    {
      "epoch": 26.03,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.0005,
      "step": 270
    },
    {
      "epoch": 26.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0010995334014296532,
      "eval_runtime": 5.1379,
      "eval_samples_per_second": 13.624,
      "eval_steps_per_second": 0.584,
      "step": 270
    },
    {
      "epoch": 27.03,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.0005,
      "step": 280
    },
    {
      "epoch": 27.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0010933494195342064,
      "eval_runtime": 5.5747,
      "eval_samples_per_second": 12.557,
      "eval_steps_per_second": 0.538,
      "step": 280
    },
    {
      "epoch": 28.03,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.0005,
      "step": 290
    },
    {
      "epoch": 28.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0010898270411416888,
      "eval_runtime": 5.3602,
      "eval_samples_per_second": 13.059,
      "eval_steps_per_second": 0.56,
      "step": 290
    },
    {
      "epoch": 29.03,
      "learning_rate": 0.0,
      "loss": 0.0005,
      "step": 300
    },
    {
      "epoch": 29.03,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0010924884118139744,
      "eval_runtime": 6.1154,
      "eval_samples_per_second": 11.447,
      "eval_steps_per_second": 0.491,
      "step": 300
    },
    {
      "epoch": 29.03,
      "step": 300,
      "total_flos": 3.952030329844531e+19,
      "train_loss": 0.2041329901261876,
      "train_runtime": 1416.8695,
      "train_samples_per_second": 6.776,
      "train_steps_per_second": 0.212
    },
    {
      "epoch": 29.03,
      "eval_accuracy": 0.9741935483870968,
      "eval_loss": 0.09911247342824936,
      "eval_runtime": 16.8335,
      "eval_samples_per_second": 9.208,
      "eval_steps_per_second": 0.297,
      "step": 300
    },
    {
      "epoch": 29.03,
      "eval_accuracy": 0.9741935483870968,
      "eval_loss": 0.09928672015666962,
      "eval_runtime": 13.0498,
      "eval_samples_per_second": 11.878,
      "eval_steps_per_second": 0.383,
      "step": 300
    }
  ],
  "logging_steps": 10,
  "max_steps": 300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 3.952030329844531e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}