trainer_state.json

{
  "best_metric": 0.8784,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0074/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 3.154751777648926,
      "learning_rate": 0.0001,
      "loss": 1.6495,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8016,
      "eval_loss": 0.7374061346054077,
      "eval_runtime": 6.7579,
      "eval_samples_per_second": 554.904,
      "eval_steps_per_second": 8.73,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.5607,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8453333333333334,
      "eval_loss": 0.523456335067749,
      "eval_runtime": 6.7865,
      "eval_samples_per_second": 552.566,
      "eval_steps_per_second": 8.694,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 25.327192306518555,
      "learning_rate": 0.0001,
      "loss": 0.3453,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8544,
      "eval_loss": 0.4795020818710327,
      "eval_runtime": 7.2404,
      "eval_samples_per_second": 517.926,
      "eval_steps_per_second": 8.149,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7294626832008362,
      "learning_rate": 0.0001,
      "loss": 0.244,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.44120317697525024,
      "eval_runtime": 6.9304,
      "eval_samples_per_second": 541.094,
      "eval_steps_per_second": 8.513,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.291212558746338,
      "learning_rate": 0.0001,
      "loss": 0.1774,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8784,
      "eval_loss": 0.440028578042984,
      "eval_runtime": 7.2733,
      "eval_samples_per_second": 515.581,
      "eval_steps_per_second": 8.112,
      "step": 1665
    }
  ],
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}