trainer_state.json

{
  "best_metric": 0.8653333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0288/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": 23.662822723388672,
      "learning_rate": 7e-05,
      "loss": 0.8727,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.844,
      "eval_loss": 0.49961337447166443,
      "eval_runtime": 7.021,
      "eval_samples_per_second": 534.11,
      "eval_steps_per_second": 8.403,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.27215030789375305,
      "learning_rate": 7e-05,
      "loss": 0.3133,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8512,
      "eval_loss": 0.488959938287735,
      "eval_runtime": 6.4951,
      "eval_samples_per_second": 577.36,
      "eval_steps_per_second": 9.084,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.013715147972107,
      "learning_rate": 7e-05,
      "loss": 0.1942,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8509333333333333,
      "eval_loss": 0.5400459170341492,
      "eval_runtime": 6.3931,
      "eval_samples_per_second": 586.568,
      "eval_steps_per_second": 9.229,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.4345707893371582,
      "learning_rate": 7e-05,
      "loss": 0.1494,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8514666666666667,
      "eval_loss": 0.5413629412651062,
      "eval_runtime": 6.3563,
      "eval_samples_per_second": 589.968,
      "eval_steps_per_second": 9.282,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.00772180175408721,
      "learning_rate": 7e-05,
      "loss": 0.1253,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.5749999284744263,
      "eval_runtime": 6.4842,
      "eval_samples_per_second": 578.33,
      "eval_steps_per_second": 9.099,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.00045733252773061395,
      "learning_rate": 7e-05,
      "loss": 0.1058,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8570666666666666,
      "eval_loss": 0.5814070105552673,
      "eval_runtime": 8.4209,
      "eval_samples_per_second": 445.322,
      "eval_steps_per_second": 7.006,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.1219635009765625,
      "learning_rate": 7e-05,
      "loss": 0.0899,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.848,
      "eval_loss": 0.6326554417610168,
      "eval_runtime": 7.0218,
      "eval_samples_per_second": 534.052,
      "eval_steps_per_second": 8.402,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.027944574132561684,
      "learning_rate": 7e-05,
      "loss": 0.083,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.5761622786521912,
      "eval_runtime": 6.2582,
      "eval_samples_per_second": 599.213,
      "eval_steps_per_second": 9.428,
      "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
}