{
  "best_metric": 0.9142857142857143,
  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-148",
  "epoch": 3.22972972972973,
  "eval_steps": 500,
  "global_step": 148,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "grad_norm": 5.877252101898193,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 2.3334,
      "step": 10
    },
    {
      "epoch": 0.14,
      "grad_norm": 6.845406532287598,
      "learning_rate": 4.81203007518797e-05,
      "loss": 2.2303,
      "step": 20
    },
    {
      "epoch": 0.2,
      "grad_norm": 8.416325569152832,
      "learning_rate": 4.43609022556391e-05,
      "loss": 2.1155,
      "step": 30
    },
    {
      "epoch": 0.26,
      "eval_accuracy": 0.6,
      "eval_loss": 1.7389822006225586,
      "eval_runtime": 7.8855,
      "eval_samples_per_second": 8.877,
      "eval_steps_per_second": 1.141,
      "step": 38
    },
    {
      "epoch": 1.01,
      "grad_norm": 10.510369300842285,
      "learning_rate": 4.0601503759398494e-05,
      "loss": 1.9057,
      "step": 40
    },
    {
      "epoch": 1.08,
      "grad_norm": 12.991025924682617,
      "learning_rate": 3.6842105263157895e-05,
      "loss": 1.4769,
      "step": 50
    },
    {
      "epoch": 1.15,
      "grad_norm": 10.709514617919922,
      "learning_rate": 3.3082706766917295e-05,
      "loss": 1.1806,
      "step": 60
    },
    {
      "epoch": 1.22,
      "grad_norm": 6.943473815917969,
      "learning_rate": 2.9323308270676693e-05,
      "loss": 0.8482,
      "step": 70
    },
    {
      "epoch": 1.26,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.954380452632904,
      "eval_runtime": 7.9006,
      "eval_samples_per_second": 8.86,
      "eval_steps_per_second": 1.139,
      "step": 76
    },
    {
      "epoch": 2.03,
      "grad_norm": 4.553641319274902,
      "learning_rate": 2.556390977443609e-05,
      "loss": 0.8814,
      "step": 80
    },
    {
      "epoch": 2.09,
      "grad_norm": 6.816486835479736,
      "learning_rate": 2.1804511278195487e-05,
      "loss": 0.6541,
      "step": 90
    },
    {
      "epoch": 2.16,
      "grad_norm": 14.017226219177246,
      "learning_rate": 1.8045112781954888e-05,
      "loss": 0.5383,
      "step": 100
    },
    {
      "epoch": 2.23,
      "grad_norm": 6.129795074462891,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.4371,
      "step": 110
    },
    {
      "epoch": 2.26,
      "eval_accuracy": 0.8857142857142857,
      "eval_loss": 0.4790053367614746,
      "eval_runtime": 7.8047,
      "eval_samples_per_second": 8.969,
      "eval_steps_per_second": 1.153,
      "step": 114
    },
    {
      "epoch": 3.04,
      "grad_norm": 4.292072772979736,
      "learning_rate": 1.0526315789473684e-05,
      "loss": 0.3274,
      "step": 120
    },
    {
      "epoch": 3.11,
      "grad_norm": 3.278257369995117,
      "learning_rate": 6.766917293233083e-06,
      "loss": 0.3116,
      "step": 130
    },
    {
      "epoch": 3.18,
      "grad_norm": 6.118241310119629,
      "learning_rate": 3.007518796992481e-06,
      "loss": 0.2803,
      "step": 140
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.9142857142857143,
      "eval_loss": 0.3659238815307617,
      "eval_runtime": 7.3677,
      "eval_samples_per_second": 9.501,
      "eval_steps_per_second": 1.222,
      "step": 148
    },
    {
      "epoch": 3.23,
      "step": 148,
      "total_flos": 1.460491890402263e+18,
      "train_loss": 1.0632461502745345,
      "train_runtime": 248.2758,
      "train_samples_per_second": 4.769,
      "train_steps_per_second": 0.596
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.8774193548387097,
      "eval_loss": 0.45682066679000854,
      "eval_runtime": 16.891,
      "eval_samples_per_second": 9.176,
      "eval_steps_per_second": 1.184,
      "step": 148
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.8774193548387097,
      "eval_loss": 0.45682066679000854,
      "eval_runtime": 15.7238,
      "eval_samples_per_second": 9.858,
      "eval_steps_per_second": 1.272,
      "step": 148
    }
  ],
  "logging_steps": 10,
  "max_steps": 148,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 1.460491890402263e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}