trainer_state.json

{
  "best_metric": 0.9,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0120/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": 55.081016540527344,
      "learning_rate": 9.621650897886543e-05,
      "loss": 0.976,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7989333333333334,
      "eval_loss": 0.6352855563163757,
      "eval_runtime": 13.1091,
      "eval_samples_per_second": 286.061,
      "eval_steps_per_second": 4.501,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8172094821929932,
      "learning_rate": 8.543862808774192e-05,
      "loss": 0.4061,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8266666666666667,
      "eval_loss": 0.5535526871681213,
      "eval_runtime": 13.7936,
      "eval_samples_per_second": 271.866,
      "eval_steps_per_second": 4.277,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.22330023348331451,
      "learning_rate": 6.924306924189557e-05,
      "loss": 0.2204,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8306666666666667,
      "eval_loss": 0.6097951531410217,
      "eval_runtime": 13.2093,
      "eval_samples_per_second": 283.891,
      "eval_steps_per_second": 4.467,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0106199374422431,
      "learning_rate": 5.011792754282769e-05,
      "loss": 0.1216,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8573333333333333,
      "eval_loss": 0.5307042002677917,
      "eval_runtime": 13.9022,
      "eval_samples_per_second": 269.741,
      "eval_steps_per_second": 4.244,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.4796805381774902,
      "learning_rate": 3.0974832444380136e-05,
      "loss": 0.0612,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.48733147978782654,
      "eval_runtime": 13.0101,
      "eval_samples_per_second": 288.237,
      "eval_steps_per_second": 4.535,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0014673619298264384,
      "learning_rate": 1.472814664270234e-05,
      "loss": 0.0185,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8877333333333334,
      "eval_loss": 0.4508797526359558,
      "eval_runtime": 13.7063,
      "eval_samples_per_second": 273.597,
      "eval_steps_per_second": 4.305,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.22747249901294708,
      "learning_rate": 3.851280774526222e-06,
      "loss": 0.0048,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.896,
      "eval_loss": 0.43265169858932495,
      "eval_runtime": 12.9714,
      "eval_samples_per_second": 289.097,
      "eval_steps_per_second": 4.548,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.000375419098418206,
      "learning_rate": 1.3906924697382196e-10,
      "loss": 0.0016,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.41788938641548157,
      "eval_runtime": 13.0832,
      "eval_samples_per_second": 286.627,
      "eval_steps_per_second": 4.51,
      "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
}