trainer_state.json

{
  "best_metric": 0.5524861878453039,
  "best_model_checkpoint": "videomae-base-finetuned-lift-data-resize/checkpoint-76",
  "epoch": 3.2105263157894735,
  "eval_steps": 500,
  "global_step": 76,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13157894736842105,
      "grad_norm": 5.10792875289917,
      "learning_rate": 4.8529411764705885e-05,
      "loss": 1.6509,
      "step": 10
    },
    {
      "epoch": 0.2631578947368421,
      "grad_norm": 5.389455795288086,
      "learning_rate": 4.11764705882353e-05,
      "loss": 1.5445,
      "step": 20
    },
    {
      "epoch": 0.2631578947368421,
      "eval_accuracy": 0.11233885819521179,
      "eval_loss": 1.6094709634780884,
      "eval_runtime": 1927.838,
      "eval_samples_per_second": 0.282,
      "eval_steps_per_second": 0.035,
      "step": 20
    },
    {
      "epoch": 1.131578947368421,
      "grad_norm": 5.348465442657471,
      "learning_rate": 3.382352941176471e-05,
      "loss": 1.3201,
      "step": 30
    },
    {
      "epoch": 1.263157894736842,
      "grad_norm": 8.077356338500977,
      "learning_rate": 2.647058823529412e-05,
      "loss": 1.2942,
      "step": 40
    },
    {
      "epoch": 1.263157894736842,
      "eval_accuracy": 0.5119705340699816,
      "eval_loss": 1.2199088335037231,
      "eval_runtime": 1935.0816,
      "eval_samples_per_second": 0.281,
      "eval_steps_per_second": 0.035,
      "step": 40
    },
    {
      "epoch": 2.1315789473684212,
      "grad_norm": 7.384807109832764,
      "learning_rate": 1.9117647058823528e-05,
      "loss": 1.0568,
      "step": 50
    },
    {
      "epoch": 2.263157894736842,
      "grad_norm": 7.400702953338623,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 1.2008,
      "step": 60
    },
    {
      "epoch": 2.263157894736842,
      "eval_accuracy": 0.5267034990791897,
      "eval_loss": 1.1846799850463867,
      "eval_runtime": 1919.9448,
      "eval_samples_per_second": 0.283,
      "eval_steps_per_second": 0.035,
      "step": 60
    },
    {
      "epoch": 3.1315789473684212,
      "grad_norm": 8.256953239440918,
      "learning_rate": 4.411764705882353e-06,
      "loss": 1.0759,
      "step": 70
    },
    {
      "epoch": 3.2105263157894735,
      "eval_accuracy": 0.5524861878453039,
      "eval_loss": 1.04552161693573,
      "eval_runtime": 1926.4299,
      "eval_samples_per_second": 0.282,
      "eval_steps_per_second": 0.035,
      "step": 76
    },
    {
      "epoch": 3.2105263157894735,
      "step": 76,
      "total_flos": 7.46412658398167e+17,
      "train_loss": 1.275582376279329,
      "train_runtime": 10041.2113,
      "train_samples_per_second": 0.061,
      "train_steps_per_second": 0.008
    },
    {
      "epoch": 3.2105263157894735,
      "eval_accuracy": 0.5407925407925408,
      "eval_loss": 0.993989884853363,
      "eval_runtime": 1508.347,
      "eval_samples_per_second": 0.284,
      "eval_steps_per_second": 0.036,
      "step": 76
    },
    {
      "epoch": 3.2105263157894735,
      "eval_accuracy": 0.5407925407925408,
      "eval_loss": 0.9939900040626526,
      "eval_runtime": 1531.0787,
      "eval_samples_per_second": 0.28,
      "eval_steps_per_second": 0.035,
      "step": 76
    }
  ],
  "logging_steps": 10,
  "max_steps": 76,
  "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": 7.46412658398167e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}