trainer_state.json

{
  "best_metric": 0.8986666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0148/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": 0.6798107028007507,
      "learning_rate": 8.753753753753754e-05,
      "loss": 0.8849,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8098666666666666,
      "eval_loss": 0.612926185131073,
      "eval_runtime": 10.5907,
      "eval_samples_per_second": 354.084,
      "eval_steps_per_second": 5.571,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7.507507507507507e-05,
      "loss": 0.3254,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8413333333333334,
      "eval_loss": 0.5488036274909973,
      "eval_runtime": 10.3369,
      "eval_samples_per_second": 362.777,
      "eval_steps_per_second": 5.708,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.4322267174720764,
      "learning_rate": 6.257507507507508e-05,
      "loss": 0.1561,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8493333333333334,
      "eval_loss": 0.5452387928962708,
      "eval_runtime": 10.4526,
      "eval_samples_per_second": 358.761,
      "eval_steps_per_second": 5.645,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5617873072624207,
      "learning_rate": 5.0075075075075074e-05,
      "loss": 0.0939,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.552000105381012,
      "eval_runtime": 10.4963,
      "eval_samples_per_second": 357.27,
      "eval_steps_per_second": 5.621,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.016121232882142067,
      "learning_rate": 3.7575075075075075e-05,
      "loss": 0.0434,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.5107405185699463,
      "eval_runtime": 10.3497,
      "eval_samples_per_second": 362.33,
      "eval_steps_per_second": 5.701,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.03188542276620865,
      "learning_rate": 2.5075075075075076e-05,
      "loss": 0.0178,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8882666666666666,
      "eval_loss": 0.4914809465408325,
      "eval_runtime": 10.8097,
      "eval_samples_per_second": 346.91,
      "eval_steps_per_second": 5.458,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.015668142586946487,
      "learning_rate": 1.2575075075075077e-05,
      "loss": 0.0047,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.4768809378147125,
      "eval_runtime": 10.8061,
      "eval_samples_per_second": 347.026,
      "eval_steps_per_second": 5.46,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.00028790938085876405,
      "learning_rate": 7.507507507507508e-08,
      "loss": 0.0008,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8986666666666666,
      "eval_loss": 0.4680449962615967,
      "eval_runtime": 10.626,
      "eval_samples_per_second": 352.907,
      "eval_steps_per_second": 5.552,
      "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
}