trainer_state.json

{
  "best_metric": 0.8573333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0978/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": Infinity,
      "learning_rate": 9e-05,
      "loss": 1.5562,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7936,
      "eval_loss": 0.7252557873725891,
      "eval_runtime": 5.5107,
      "eval_samples_per_second": 680.498,
      "eval_steps_per_second": 10.707,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5383970141410828,
      "learning_rate": 9e-05,
      "loss": 0.4687,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.836,
      "eval_loss": 0.5621740818023682,
      "eval_runtime": 6.4776,
      "eval_samples_per_second": 578.915,
      "eval_steps_per_second": 9.108,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.1410586833953857,
      "learning_rate": 9e-05,
      "loss": 0.2499,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8370666666666666,
      "eval_loss": 0.5582749843597412,
      "eval_runtime": 7.9931,
      "eval_samples_per_second": 469.154,
      "eval_steps_per_second": 7.381,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.43022504448890686,
      "learning_rate": 9e-05,
      "loss": 0.1504,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8482666666666666,
      "eval_loss": 0.5565549731254578,
      "eval_runtime": 7.7388,
      "eval_samples_per_second": 484.574,
      "eval_steps_per_second": 7.624,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.2985092401504517,
      "learning_rate": 9e-05,
      "loss": 0.1116,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8573333333333333,
      "eval_loss": 0.574180543422699,
      "eval_runtime": 6.0177,
      "eval_samples_per_second": 623.159,
      "eval_steps_per_second": 9.804,
      "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
}