{
  "best_metric": 0.9,
  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-375",
  "epoch": 4.2,
  "eval_steps": 500,
  "global_step": 375,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "grad_norm": 10.424443244934082,
      "learning_rate": 1.3157894736842106e-05,
      "loss": 2.3451,
      "step": 10
    },
    {
      "epoch": 0.05,
      "grad_norm": 9.281598091125488,
      "learning_rate": 2.6315789473684212e-05,
      "loss": 2.3445,
      "step": 20
    },
    {
      "epoch": 0.08,
      "grad_norm": 9.415879249572754,
      "learning_rate": 3.9473684210526316e-05,
      "loss": 2.234,
      "step": 30
    },
    {
      "epoch": 0.11,
      "grad_norm": 9.546265602111816,
      "learning_rate": 4.9703264094955494e-05,
      "loss": 2.1348,
      "step": 40
    },
    {
      "epoch": 0.13,
      "grad_norm": 12.744800567626953,
      "learning_rate": 4.821958456973294e-05,
      "loss": 2.0604,
      "step": 50
    },
    {
      "epoch": 0.16,
      "grad_norm": 9.805941581726074,
      "learning_rate": 4.673590504451038e-05,
      "loss": 1.8655,
      "step": 60
    },
    {
      "epoch": 0.19,
      "grad_norm": 15.323193550109863,
      "learning_rate": 4.525222551928784e-05,
      "loss": 1.9591,
      "step": 70
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.42857142857142855,
      "eval_loss": 1.6628010272979736,
      "eval_runtime": 13.2129,
      "eval_samples_per_second": 5.298,
      "eval_steps_per_second": 1.362,
      "step": 75
    },
    {
      "epoch": 1.01,
      "grad_norm": 11.818169593811035,
      "learning_rate": 4.3768545994065286e-05,
      "loss": 1.482,
      "step": 80
    },
    {
      "epoch": 1.04,
      "grad_norm": 12.213088035583496,
      "learning_rate": 4.228486646884273e-05,
      "loss": 1.2242,
      "step": 90
    },
    {
      "epoch": 1.07,
      "grad_norm": 18.863269805908203,
      "learning_rate": 4.080118694362018e-05,
      "loss": 1.0472,
      "step": 100
    },
    {
      "epoch": 1.09,
      "grad_norm": 10.183364868164062,
      "learning_rate": 3.9317507418397627e-05,
      "loss": 0.7998,
      "step": 110
    },
    {
      "epoch": 1.12,
      "grad_norm": 9.917871475219727,
      "learning_rate": 3.783382789317508e-05,
      "loss": 0.837,
      "step": 120
    },
    {
      "epoch": 1.15,
      "grad_norm": 15.699859619140625,
      "learning_rate": 3.635014836795252e-05,
      "loss": 0.6737,
      "step": 130
    },
    {
      "epoch": 1.17,
      "grad_norm": 11.725618362426758,
      "learning_rate": 3.4866468842729974e-05,
      "loss": 0.6864,
      "step": 140
    },
    {
      "epoch": 1.2,
      "grad_norm": 65.09258270263672,
      "learning_rate": 3.338278931750742e-05,
      "loss": 0.9775,
      "step": 150
    },
    {
      "epoch": 1.2,
      "eval_accuracy": 0.7285714285714285,
      "eval_loss": 0.9276337623596191,
      "eval_runtime": 12.7097,
      "eval_samples_per_second": 5.508,
      "eval_steps_per_second": 1.416,
      "step": 150
    },
    {
      "epoch": 2.03,
      "grad_norm": 11.323193550109863,
      "learning_rate": 3.189910979228487e-05,
      "loss": 0.4997,
      "step": 160
    },
    {
      "epoch": 2.05,
      "grad_norm": 34.906097412109375,
      "learning_rate": 3.0415430267062318e-05,
      "loss": 0.5372,
      "step": 170
    },
    {
      "epoch": 2.08,
      "grad_norm": 4.410595893859863,
      "learning_rate": 2.8931750741839762e-05,
      "loss": 0.5236,
      "step": 180
    },
    {
      "epoch": 2.11,
      "grad_norm": 11.69422721862793,
      "learning_rate": 2.744807121661721e-05,
      "loss": 0.3245,
      "step": 190
    },
    {
      "epoch": 2.13,
      "grad_norm": 16.422433853149414,
      "learning_rate": 2.5964391691394662e-05,
      "loss": 0.4124,
      "step": 200
    },
    {
      "epoch": 2.16,
      "grad_norm": 5.309849739074707,
      "learning_rate": 2.4480712166172106e-05,
      "loss": 0.4554,
      "step": 210
    },
    {
      "epoch": 2.19,
      "grad_norm": 1.0133864879608154,
      "learning_rate": 2.2997032640949558e-05,
      "loss": 0.2693,
      "step": 220
    },
    {
      "epoch": 2.2,
      "eval_accuracy": 0.7428571428571429,
      "eval_loss": 0.4798918664455414,
      "eval_runtime": 12.6824,
      "eval_samples_per_second": 5.519,
      "eval_steps_per_second": 1.419,
      "step": 225
    },
    {
      "epoch": 3.01,
      "grad_norm": 1.2999351024627686,
      "learning_rate": 2.1513353115727002e-05,
      "loss": 0.2994,
      "step": 230
    },
    {
      "epoch": 3.04,
      "grad_norm": 0.5214590430259705,
      "learning_rate": 2.0029673590504454e-05,
      "loss": 0.0871,
      "step": 240
    },
    {
      "epoch": 3.07,
      "grad_norm": 22.973987579345703,
      "learning_rate": 1.85459940652819e-05,
      "loss": 0.1398,
      "step": 250
    },
    {
      "epoch": 3.09,
      "grad_norm": 0.22316251695156097,
      "learning_rate": 1.706231454005935e-05,
      "loss": 0.0745,
      "step": 260
    },
    {
      "epoch": 3.12,
      "grad_norm": 1.7083332538604736,
      "learning_rate": 1.5578635014836794e-05,
      "loss": 0.2018,
      "step": 270
    },
    {
      "epoch": 3.15,
      "grad_norm": 0.34350940585136414,
      "learning_rate": 1.4094955489614246e-05,
      "loss": 0.282,
      "step": 280
    },
    {
      "epoch": 3.17,
      "grad_norm": 2.1111364364624023,
      "learning_rate": 1.2611275964391692e-05,
      "loss": 0.2336,
      "step": 290
    },
    {
      "epoch": 3.2,
      "grad_norm": 2.993694305419922,
      "learning_rate": 1.112759643916914e-05,
      "loss": 0.2194,
      "step": 300
    },
    {
      "epoch": 3.2,
      "eval_accuracy": 0.8285714285714286,
      "eval_loss": 0.43333032727241516,
      "eval_runtime": 12.7052,
      "eval_samples_per_second": 5.51,
      "eval_steps_per_second": 1.417,
      "step": 300
    },
    {
      "epoch": 4.03,
      "grad_norm": 0.44264498353004456,
      "learning_rate": 9.643916913946588e-06,
      "loss": 0.0751,
      "step": 310
    },
    {
      "epoch": 4.05,
      "grad_norm": 8.502839088439941,
      "learning_rate": 8.160237388724036e-06,
      "loss": 0.2655,
      "step": 320
    },
    {
      "epoch": 4.08,
      "grad_norm": 11.5534086227417,
      "learning_rate": 6.676557863501484e-06,
      "loss": 0.2037,
      "step": 330
    },
    {
      "epoch": 4.11,
      "grad_norm": 1.2078895568847656,
      "learning_rate": 5.192878338278932e-06,
      "loss": 0.0657,
      "step": 340
    },
    {
      "epoch": 4.13,
      "grad_norm": 0.2843732535839081,
      "learning_rate": 3.7091988130563796e-06,
      "loss": 0.1069,
      "step": 350
    },
    {
      "epoch": 4.16,
      "grad_norm": 0.15084530413150787,
      "learning_rate": 2.225519287833828e-06,
      "loss": 0.073,
      "step": 360
    },
    {
      "epoch": 4.19,
      "grad_norm": 0.18272823095321655,
      "learning_rate": 7.41839762611276e-07,
      "loss": 0.0873,
      "step": 370
    },
    {
      "epoch": 4.2,
      "eval_accuracy": 0.9,
      "eval_loss": 0.31446826457977295,
      "eval_runtime": 12.7219,
      "eval_samples_per_second": 5.502,
      "eval_steps_per_second": 1.415,
      "step": 375
    },
    {
      "epoch": 4.2,
      "step": 375,
      "total_flos": 1.86923023515648e+18,
      "train_loss": 0.7497338762482008,
      "train_runtime": 612.0494,
      "train_samples_per_second": 2.451,
      "train_steps_per_second": 0.613
    },
    {
      "epoch": 4.2,
      "eval_accuracy": 0.9161290322580645,
      "eval_loss": 0.25597551465034485,
      "eval_runtime": 29.0097,
      "eval_samples_per_second": 5.343,
      "eval_steps_per_second": 1.344,
      "step": 375
    },
    {
      "epoch": 4.2,
      "eval_accuracy": 0.9161290322580645,
      "eval_loss": 0.25597548484802246,
      "eval_runtime": 28.3145,
      "eval_samples_per_second": 5.474,
      "eval_steps_per_second": 1.377,
      "step": 375
    }
  ],
  "logging_steps": 10,
  "max_steps": 375,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 1.86923023515648e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}