trainer_state.json

{
  "best_metric": 0.44293333333333335,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0216/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 20.89984893798828,
      "learning_rate": 0.0003,
      "loss": 3.3031,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.23946666666666666,
      "eval_loss": 2.95447039604187,
      "eval_runtime": 6.8257,
      "eval_samples_per_second": 549.393,
      "eval_steps_per_second": 8.644,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 24.805265426635742,
      "learning_rate": 0.0003,
      "loss": 2.7381,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.28613333333333335,
      "eval_loss": 2.6918601989746094,
      "eval_runtime": 6.6789,
      "eval_samples_per_second": 561.467,
      "eval_steps_per_second": 8.834,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 21.222753524780273,
      "learning_rate": 0.0003,
      "loss": 2.4726,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.32986666666666664,
      "eval_loss": 2.4872984886169434,
      "eval_runtime": 6.6943,
      "eval_samples_per_second": 560.182,
      "eval_steps_per_second": 8.814,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 25.330617904663086,
      "learning_rate": 0.0003,
      "loss": 2.2655,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.36506666666666665,
      "eval_loss": 2.3660078048706055,
      "eval_runtime": 6.852,
      "eval_samples_per_second": 547.286,
      "eval_steps_per_second": 8.611,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 21.920289993286133,
      "learning_rate": 0.0003,
      "loss": 2.0679,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3896,
      "eval_loss": 2.2779324054718018,
      "eval_runtime": 6.7632,
      "eval_samples_per_second": 554.474,
      "eval_steps_per_second": 8.724,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 24.18760871887207,
      "learning_rate": 0.0003,
      "loss": 1.894,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.41333333333333333,
      "eval_loss": 2.2070207595825195,
      "eval_runtime": 7.3001,
      "eval_samples_per_second": 513.694,
      "eval_steps_per_second": 8.082,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 1.7273,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4256,
      "eval_loss": 2.123703718185425,
      "eval_runtime": 7.4332,
      "eval_samples_per_second": 504.495,
      "eval_steps_per_second": 7.937,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 35.57888412475586,
      "learning_rate": 0.0003,
      "loss": 1.5629,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.44293333333333335,
      "eval_loss": 2.101738691329956,
      "eval_runtime": 6.9798,
      "eval_samples_per_second": 537.265,
      "eval_steps_per_second": 8.453,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}