trainer_state.json

{
  "best_metric": 0.9018666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0957/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": 59.57599639892578,
      "learning_rate": 6.790177189676492e-05,
      "loss": 0.8753,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8232,
      "eval_loss": 0.5857082009315491,
      "eval_runtime": 8.5402,
      "eval_samples_per_second": 439.101,
      "eval_steps_per_second": 6.909,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.16090914607048035,
      "learning_rate": 6.183512375367748e-05,
      "loss": 0.3134,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8461333333333333,
      "eval_loss": 0.505794882774353,
      "eval_runtime": 5.9761,
      "eval_samples_per_second": 627.504,
      "eval_steps_per_second": 9.873,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 19.401187896728516,
      "learning_rate": 5.2563508019062023e-05,
      "loss": 0.162,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8432,
      "eval_loss": 0.5433351397514343,
      "eval_runtime": 5.968,
      "eval_samples_per_second": 628.346,
      "eval_steps_per_second": 9.886,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.748925507068634,
      "learning_rate": 4.1149935250010576e-05,
      "loss": 0.0847,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8661333333333333,
      "eval_loss": 0.498929500579834,
      "eval_runtime": 5.9822,
      "eval_samples_per_second": 626.863,
      "eval_steps_per_second": 9.863,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0022496578749269247,
      "learning_rate": 2.8994589526430138e-05,
      "loss": 0.0475,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8824,
      "eval_loss": 0.5032458305358887,
      "eval_runtime": 6.0171,
      "eval_samples_per_second": 623.224,
      "eval_steps_per_second": 9.805,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.008986296132206917,
      "learning_rate": 1.756358493638142e-05,
      "loss": 0.0178,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8896,
      "eval_loss": 0.4813222587108612,
      "eval_runtime": 5.9894,
      "eval_samples_per_second": 626.102,
      "eval_steps_per_second": 9.851,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0008834514301270247,
      "learning_rate": 8.235669337084578e-06,
      "loss": 0.0054,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8986666666666666,
      "eval_loss": 0.42193105816841125,
      "eval_runtime": 6.1728,
      "eval_samples_per_second": 607.503,
      "eval_steps_per_second": 9.558,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0003348967875353992,
      "learning_rate": 2.135927015187301e-06,
      "loss": 0.002,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9018666666666667,
      "eval_loss": 0.4154098331928253,
      "eval_runtime": 5.983,
      "eval_samples_per_second": 626.776,
      "eval_steps_per_second": 9.861,
      "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
}