trainer_state.json

{
  "best_metric": 0.9010666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0224/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 9.621650897886543e-05,
      "loss": 1.7262,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7845333333333333,
      "eval_loss": 0.7970544695854187,
      "eval_runtime": 12.4694,
      "eval_samples_per_second": 300.737,
      "eval_steps_per_second": 4.732,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.189880132675171,
      "learning_rate": 8.539700818671957e-05,
      "loss": 0.569,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8194666666666667,
      "eval_loss": 0.635564386844635,
      "eval_runtime": 11.2398,
      "eval_samples_per_second": 333.637,
      "eval_steps_per_second": 5.249,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 16.50010108947754,
      "learning_rate": 6.918863377282392e-05,
      "loss": 0.3539,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8618666666666667,
      "eval_loss": 0.4697021543979645,
      "eval_runtime": 11.2844,
      "eval_samples_per_second": 332.317,
      "eval_steps_per_second": 5.228,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7012445330619812,
      "learning_rate": 5.005896381241413e-05,
      "loss": 0.2033,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8770666666666667,
      "eval_loss": 0.4208700656890869,
      "eval_runtime": 11.2638,
      "eval_samples_per_second": 332.926,
      "eval_steps_per_second": 5.238,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.07219620048999786,
      "learning_rate": 3.0920317146072576e-05,
      "loss": 0.1105,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8866666666666667,
      "eval_loss": 0.4118160903453827,
      "eval_runtime": 11.1994,
      "eval_samples_per_second": 334.839,
      "eval_steps_per_second": 5.268,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 7.919136047363281,
      "learning_rate": 1.4686379236485708e-05,
      "loss": 0.0551,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9010666666666667,
      "eval_loss": 0.3738054633140564,
      "eval_runtime": 11.1306,
      "eval_samples_per_second": 336.91,
      "eval_steps_per_second": 5.301,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}