trainer_state.json

{
  "best_metric": 0.4376,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0715/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": 17.579166412353516,
      "learning_rate": 0.0004382507507507508,
      "loss": 3.358,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.1928,
      "eval_loss": 3.017362594604492,
      "eval_runtime": 11.7589,
      "eval_samples_per_second": 318.909,
      "eval_steps_per_second": 5.017,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 22.820676803588867,
      "learning_rate": 0.0003757507507507508,
      "loss": 2.8424,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.25066666666666665,
      "eval_loss": 2.779968023300171,
      "eval_runtime": 11.9559,
      "eval_samples_per_second": 313.654,
      "eval_steps_per_second": 4.935,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 23.12680435180664,
      "learning_rate": 0.0003132507507507508,
      "loss": 2.619,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3,
      "eval_loss": 2.55679988861084,
      "eval_runtime": 11.2845,
      "eval_samples_per_second": 332.314,
      "eval_steps_per_second": 5.228,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 20.347064971923828,
      "learning_rate": 0.0002507507507507508,
      "loss": 2.4298,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.33093333333333336,
      "eval_loss": 2.4048681259155273,
      "eval_runtime": 10.9376,
      "eval_samples_per_second": 342.853,
      "eval_steps_per_second": 5.394,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 18.868165969848633,
      "learning_rate": 0.00018825075075075076,
      "loss": 2.256,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3682666666666667,
      "eval_loss": 2.283622980117798,
      "eval_runtime": 10.7555,
      "eval_samples_per_second": 348.657,
      "eval_steps_per_second": 5.486,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 26.03997039794922,
      "learning_rate": 0.00012575075075075076,
      "loss": 2.0773,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3914666666666667,
      "eval_loss": 2.186504364013672,
      "eval_runtime": 10.8347,
      "eval_samples_per_second": 346.11,
      "eval_steps_per_second": 5.445,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": Infinity,
      "learning_rate": 6.343843843843844e-05,
      "loss": 1.8698,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.41706666666666664,
      "eval_loss": 2.0840203762054443,
      "eval_runtime": 10.9946,
      "eval_samples_per_second": 341.077,
      "eval_steps_per_second": 5.366,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 24.889787673950195,
      "learning_rate": 9.384384384384384e-07,
      "loss": 1.6415,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4376,
      "eval_loss": 2.0485236644744873,
      "eval_runtime": 13.1499,
      "eval_samples_per_second": 285.173,
      "eval_steps_per_second": 4.487,
      "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
}