trainer_state.json

{
  "best_metric": 0.8608,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0678/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": 1.9672043323516846,
      "learning_rate": 0.0001,
      "loss": 1.46,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7773333333333333,
      "eval_loss": 0.7602531313896179,
      "eval_runtime": 12.4064,
      "eval_samples_per_second": 302.263,
      "eval_steps_per_second": 4.756,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.38733434677124,
      "learning_rate": 0.0001,
      "loss": 0.5002,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8154666666666667,
      "eval_loss": 0.6024943590164185,
      "eval_runtime": 12.4744,
      "eval_samples_per_second": 300.615,
      "eval_steps_per_second": 4.73,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.053205490112305,
      "learning_rate": 0.0001,
      "loss": 0.3325,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8458666666666667,
      "eval_loss": 0.50677090883255,
      "eval_runtime": 12.3612,
      "eval_samples_per_second": 303.368,
      "eval_steps_per_second": 4.773,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.2326,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8541333333333333,
      "eval_loss": 0.47826945781707764,
      "eval_runtime": 12.4473,
      "eval_samples_per_second": 301.27,
      "eval_steps_per_second": 4.74,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.8711986541748047,
      "learning_rate": 0.0001,
      "loss": 0.1822,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8522666666666666,
      "eval_loss": 0.5180644989013672,
      "eval_runtime": 12.2161,
      "eval_samples_per_second": 306.973,
      "eval_steps_per_second": 4.83,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.04465409368276596,
      "learning_rate": 0.0001,
      "loss": 0.133,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.5079869627952576,
      "eval_runtime": 12.1753,
      "eval_samples_per_second": 308.0,
      "eval_steps_per_second": 4.846,
      "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
}