{
  "best_metric": 0.5891016200294551,
  "best_model_checkpoint": "videomae-base-finetuned-lift-data-resize/checkpoint-140",
  "epoch": 7.102564102564102,
  "eval_steps": 500,
  "global_step": 156,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0641025641025641,
      "grad_norm": 5.635838508605957,
      "learning_rate": 3.125e-05,
      "loss": 1.6404,
      "step": 10
    },
    {
      "epoch": 0.1282051282051282,
      "grad_norm": 5.812517166137695,
      "learning_rate": 4.8571428571428576e-05,
      "loss": 1.5758,
      "step": 20
    },
    {
      "epoch": 0.1282051282051282,
      "eval_accuracy": 0.29013254786450665,
      "eval_loss": 1.6618727445602417,
      "eval_runtime": 2083.0077,
      "eval_samples_per_second": 0.326,
      "eval_steps_per_second": 0.041,
      "step": 20
    },
    {
      "epoch": 1.064102564102564,
      "grad_norm": 4.693382740020752,
      "learning_rate": 4.5e-05,
      "loss": 1.3588,
      "step": 30
    },
    {
      "epoch": 1.1282051282051282,
      "grad_norm": 7.194770336151123,
      "learning_rate": 4.1428571428571437e-05,
      "loss": 1.3067,
      "step": 40
    },
    {
      "epoch": 1.1282051282051282,
      "eval_accuracy": 0.29896907216494845,
      "eval_loss": 1.6047614812850952,
      "eval_runtime": 2110.0361,
      "eval_samples_per_second": 0.322,
      "eval_steps_per_second": 0.04,
      "step": 40
    },
    {
      "epoch": 2.064102564102564,
      "grad_norm": 9.944411277770996,
      "learning_rate": 3.785714285714286e-05,
      "loss": 1.1955,
      "step": 50
    },
    {
      "epoch": 2.128205128205128,
      "grad_norm": 8.486799240112305,
      "learning_rate": 3.428571428571429e-05,
      "loss": 1.3787,
      "step": 60
    },
    {
      "epoch": 2.128205128205128,
      "eval_accuracy": 0.3181148748159057,
      "eval_loss": 1.4723353385925293,
      "eval_runtime": 2076.9142,
      "eval_samples_per_second": 0.327,
      "eval_steps_per_second": 0.041,
      "step": 60
    },
    {
      "epoch": 3.064102564102564,
      "grad_norm": 10.37818717956543,
      "learning_rate": 3.071428571428572e-05,
      "loss": 1.2217,
      "step": 70
    },
    {
      "epoch": 3.128205128205128,
      "grad_norm": 11.789801597595215,
      "learning_rate": 2.714285714285714e-05,
      "loss": 1.1642,
      "step": 80
    },
    {
      "epoch": 3.128205128205128,
      "eval_accuracy": 0.3004418262150221,
      "eval_loss": 1.4191110134124756,
      "eval_runtime": 2171.4394,
      "eval_samples_per_second": 0.313,
      "eval_steps_per_second": 0.039,
      "step": 80
    },
    {
      "epoch": 4.064102564102564,
      "grad_norm": 4.555699348449707,
      "learning_rate": 2.357142857142857e-05,
      "loss": 1.0205,
      "step": 90
    },
    {
      "epoch": 4.128205128205128,
      "grad_norm": 7.893691539764404,
      "learning_rate": 2e-05,
      "loss": 1.1172,
      "step": 100
    },
    {
      "epoch": 4.128205128205128,
      "eval_accuracy": 0.31958762886597936,
      "eval_loss": 1.2374299764633179,
      "eval_runtime": 2014.151,
      "eval_samples_per_second": 0.337,
      "eval_steps_per_second": 0.042,
      "step": 100
    },
    {
      "epoch": 5.064102564102564,
      "grad_norm": 4.984745979309082,
      "learning_rate": 1.642857142857143e-05,
      "loss": 1.0003,
      "step": 110
    },
    {
      "epoch": 5.128205128205128,
      "grad_norm": 9.872081756591797,
      "learning_rate": 1.2857142857142857e-05,
      "loss": 0.8982,
      "step": 120
    },
    {
      "epoch": 5.128205128205128,
      "eval_accuracy": 0.5655375552282769,
      "eval_loss": 1.0099185705184937,
      "eval_runtime": 2071.3572,
      "eval_samples_per_second": 0.328,
      "eval_steps_per_second": 0.041,
      "step": 120
    },
    {
      "epoch": 6.064102564102564,
      "grad_norm": 8.504035949707031,
      "learning_rate": 9.285714285714286e-06,
      "loss": 0.8094,
      "step": 130
    },
    {
      "epoch": 6.128205128205128,
      "grad_norm": 15.16745376586914,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 0.915,
      "step": 140
    },
    {
      "epoch": 6.128205128205128,
      "eval_accuracy": 0.5891016200294551,
      "eval_loss": 0.9540217518806458,
      "eval_runtime": 2043.7965,
      "eval_samples_per_second": 0.332,
      "eval_steps_per_second": 0.042,
      "step": 140
    },
    {
      "epoch": 7.064102564102564,
      "grad_norm": 7.420783519744873,
      "learning_rate": 2.142857142857143e-06,
      "loss": 0.7809,
      "step": 150
    },
    {
      "epoch": 7.102564102564102,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 0.9188718795776367,
      "eval_runtime": 2011.7001,
      "eval_samples_per_second": 0.338,
      "eval_steps_per_second": 0.042,
      "step": 156
    },
    {
      "epoch": 7.102564102564102,
      "step": 156,
      "total_flos": 1.528962156685394e+18,
      "train_loss": 1.1473380480057154,
      "train_runtime": 20559.8951,
      "train_samples_per_second": 0.061,
      "train_steps_per_second": 0.008
    },
    {
      "epoch": 7.102564102564102,
      "eval_accuracy": 0.7680890538033395,
      "eval_loss": 0.8145598769187927,
      "eval_runtime": 1580.5762,
      "eval_samples_per_second": 0.341,
      "eval_steps_per_second": 0.043,
      "step": 156
    },
    {
      "epoch": 7.102564102564102,
      "eval_accuracy": 0.7680890538033395,
      "eval_loss": 0.8150683641433716,
      "eval_runtime": 1611.9221,
      "eval_samples_per_second": 0.334,
      "eval_steps_per_second": 0.042,
      "step": 156
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "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": 1.528962156685394e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}