trainer_state.json

{
  "best_metric": 0.9237333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0269/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": 5.210585594177246,
      "learning_rate": 4.8108254489432715e-05,
      "loss": 0.7947,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8549333333333333,
      "eval_loss": 0.46753448247909546,
      "eval_runtime": 7.1023,
      "eval_samples_per_second": 528.0,
      "eval_steps_per_second": 8.307,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5951310396194458,
      "learning_rate": 4.2698504093359784e-05,
      "loss": 0.2796,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8818666666666667,
      "eval_loss": 0.37985679507255554,
      "eval_runtime": 7.2206,
      "eval_samples_per_second": 519.349,
      "eval_steps_per_second": 8.171,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 3.4621534620947785e-05,
      "loss": 0.152,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8834666666666666,
      "eval_loss": 0.3930128812789917,
      "eval_runtime": 7.2919,
      "eval_samples_per_second": 514.266,
      "eval_steps_per_second": 8.091,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.016271529719233513,
      "learning_rate": 2.5058963771413844e-05,
      "loss": 0.0862,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8962666666666667,
      "eval_loss": 0.3567795157432556,
      "eval_runtime": 7.9995,
      "eval_samples_per_second": 468.779,
      "eval_steps_per_second": 7.375,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.01584867760539055,
      "learning_rate": 1.5487416222190068e-05,
      "loss": 0.042,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9152,
      "eval_loss": 0.3174085319042206,
      "eval_runtime": 7.1323,
      "eval_samples_per_second": 525.777,
      "eval_steps_per_second": 8.272,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0706920325756073,
      "learning_rate": 7.36407332135117e-06,
      "loss": 0.0139,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.92,
      "eval_loss": 0.288913756608963,
      "eval_runtime": 7.1366,
      "eval_samples_per_second": 525.464,
      "eval_steps_per_second": 8.267,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0249465424567461,
      "learning_rate": 1.925640387263111e-06,
      "loss": 0.0048,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9221333333333334,
      "eval_loss": 0.2913636863231659,
      "eval_runtime": 7.7612,
      "eval_samples_per_second": 483.173,
      "eval_steps_per_second": 7.602,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.006437409203499556,
      "learning_rate": 6.953462348691098e-11,
      "loss": 0.0021,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9237333333333333,
      "eval_loss": 0.29266300797462463,
      "eval_runtime": 7.1925,
      "eval_samples_per_second": 521.375,
      "eval_steps_per_second": 8.203,
      "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
}