trainer_state.json

{
  "best_metric": 0.9221333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0036/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 25.133424758911133,
      "learning_rate": 4.528073362367401e-05,
      "loss": 0.7831,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8709333333333333,
      "eval_loss": 0.4172894060611725,
      "eval_runtime": 12.3159,
      "eval_samples_per_second": 304.485,
      "eval_steps_per_second": 4.791,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 17.16738510131836,
      "learning_rate": 3.281509432839476e-05,
      "loss": 0.2067,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8858666666666667,
      "eval_loss": 0.39839667081832886,
      "eval_runtime": 12.4602,
      "eval_samples_per_second": 300.959,
      "eval_steps_per_second": 4.735,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.9162890315055847,
      "learning_rate": 1.736435462495525e-05,
      "loss": 0.0702,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9072,
      "eval_loss": 0.34446537494659424,
      "eval_runtime": 12.3794,
      "eval_samples_per_second": 302.922,
      "eval_steps_per_second": 4.766,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3632631301879883,
      "learning_rate": 4.8301719287419e-06,
      "loss": 0.016,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9184,
      "eval_loss": 0.29222407937049866,
      "eval_runtime": 12.8938,
      "eval_samples_per_second": 290.837,
      "eval_steps_per_second": 4.576,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.009306316263973713,
      "learning_rate": 1.7800850740101738e-10,
      "loss": 0.003,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9221333333333334,
      "eval_loss": 0.27743014693260193,
      "eval_runtime": 12.4215,
      "eval_samples_per_second": 301.897,
      "eval_steps_per_second": 4.75,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}