trainer_state.json

{
  "best_metric": 0.44666666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0219/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": 27.751188278198242,
      "learning_rate": 0.0002911677661332091,
      "loss": 3.369,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2144,
      "eval_loss": 2.95810604095459,
      "eval_runtime": 7.4236,
      "eval_samples_per_second": 505.143,
      "eval_steps_per_second": 7.948,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 29.241426467895508,
      "learning_rate": 0.0002653099348375843,
      "loss": 2.8313,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.26293333333333335,
      "eval_loss": 2.72469425201416,
      "eval_runtime": 8.5587,
      "eval_samples_per_second": 438.15,
      "eval_steps_per_second": 6.894,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 24.419740676879883,
      "learning_rate": 0.00022554402360715464,
      "loss": 2.5845,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3272,
      "eval_loss": 2.5356192588806152,
      "eval_runtime": 9.329,
      "eval_samples_per_second": 401.972,
      "eval_steps_per_second": 6.324,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 27.23259735107422,
      "learning_rate": 0.00017666638821865508,
      "loss": 2.3704,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3336,
      "eval_loss": 2.448772668838501,
      "eval_runtime": 8.9167,
      "eval_samples_per_second": 420.557,
      "eval_steps_per_second": 6.617,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 14.431060791015625,
      "learning_rate": 0.00012457239285701023,
      "loss": 2.1629,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.38426666666666665,
      "eval_loss": 2.252300977706909,
      "eval_runtime": 7.4129,
      "eval_samples_per_second": 505.878,
      "eval_steps_per_second": 7.959,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 47.83271026611328,
      "learning_rate": 7.554534218832392e-05,
      "loss": 1.9352,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4106666666666667,
      "eval_loss": 2.1641783714294434,
      "eval_runtime": 7.3797,
      "eval_samples_per_second": 508.148,
      "eval_steps_per_second": 7.995,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 56.30943298339844,
      "learning_rate": 3.5498622085445925e-05,
      "loss": 1.6719,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4338666666666667,
      "eval_loss": 2.1362504959106445,
      "eval_runtime": 8.7587,
      "eval_samples_per_second": 428.147,
      "eval_steps_per_second": 6.736,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": Infinity,
      "learning_rate": 9.31693256552663e-06,
      "loss": 1.4433,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.44666666666666666,
      "eval_loss": 2.1503353118896484,
      "eval_runtime": 7.7014,
      "eval_samples_per_second": 486.924,
      "eval_steps_per_second": 7.661,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}