trainer_state.json

{
  "best_metric": 0.9237333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0345/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": 2.8013902315014018e-05,
      "loss": 0.7779,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8736,
      "eval_loss": 0.3935932219028473,
      "eval_runtime": 12.0657,
      "eval_samples_per_second": 310.799,
      "eval_steps_per_second": 4.89,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.693190097808838,
      "learning_rate": 2.2540814184580827e-05,
      "loss": 0.1517,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9085333333333333,
      "eval_loss": 0.3162733018398285,
      "eval_runtime": 12.0858,
      "eval_samples_per_second": 310.282,
      "eval_steps_per_second": 4.882,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.02267720364034176,
      "learning_rate": 1.5047170983116048e-05,
      "loss": 0.0452,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9077333333333333,
      "eval_loss": 0.3270476460456848,
      "eval_runtime": 11.9833,
      "eval_samples_per_second": 312.934,
      "eval_steps_per_second": 4.923,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.13439179956912994,
      "learning_rate": 7.5408883548191384e-06,
      "loss": 0.012,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9202666666666667,
      "eval_loss": 0.29472681879997253,
      "eval_runtime": 12.7675,
      "eval_samples_per_second": 293.714,
      "eval_steps_per_second": 4.621,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.002614903962239623,
      "learning_rate": 2.0332686681020267e-06,
      "loss": 0.0017,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9210666666666667,
      "eval_loss": 0.2955109775066376,
      "eval_runtime": 12.0139,
      "eval_samples_per_second": 312.139,
      "eval_steps_per_second": 4.911,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0024122116155922413,
      "learning_rate": 7.417023831113313e-11,
      "loss": 0.0005,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9237333333333333,
      "eval_loss": 0.2938252091407776,
      "eval_runtime": 12.465,
      "eval_samples_per_second": 300.843,
      "eval_steps_per_second": 4.733,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}