trainer_state.json

{
  "best_metric": 0.9226666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0456/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": 0.44125252962112427,
      "learning_rate": 2.626126126126126e-05,
      "loss": 0.8292,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8770666666666667,
      "eval_loss": 0.3994003236293793,
      "eval_runtime": 8.0047,
      "eval_samples_per_second": 468.475,
      "eval_steps_per_second": 7.371,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 2.2522522522522523e-05,
      "loss": 0.1951,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9002666666666667,
      "eval_loss": 0.33158212900161743,
      "eval_runtime": 6.9993,
      "eval_samples_per_second": 535.766,
      "eval_steps_per_second": 8.429,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 79.73345184326172,
      "learning_rate": 1.8772522522522523e-05,
      "loss": 0.0784,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9010666666666667,
      "eval_loss": 0.3366839289665222,
      "eval_runtime": 6.8329,
      "eval_samples_per_second": 548.817,
      "eval_steps_per_second": 8.635,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.05836740881204605,
      "learning_rate": 1.5022522522522522e-05,
      "loss": 0.0312,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.912,
      "eval_loss": 0.3271813690662384,
      "eval_runtime": 6.1776,
      "eval_samples_per_second": 607.03,
      "eval_steps_per_second": 9.551,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.002857549348846078,
      "learning_rate": 1.1272522522522524e-05,
      "loss": 0.0123,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9157333333333333,
      "eval_loss": 0.34330692887306213,
      "eval_runtime": 6.6272,
      "eval_samples_per_second": 565.849,
      "eval_steps_per_second": 8.903,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.005564517807215452,
      "learning_rate": 7.522522522522523e-06,
      "loss": 0.0046,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9170666666666667,
      "eval_loss": 0.3509453237056732,
      "eval_runtime": 5.971,
      "eval_samples_per_second": 628.031,
      "eval_steps_per_second": 9.881,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.015467842109501362,
      "learning_rate": 3.772522522522523e-06,
      "loss": 0.0025,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9208,
      "eval_loss": 0.35507819056510925,
      "eval_runtime": 6.0774,
      "eval_samples_per_second": 617.036,
      "eval_steps_per_second": 9.708,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.1553248167037964,
      "learning_rate": 2.2522522522522524e-08,
      "loss": 0.001,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9226666666666666,
      "eval_loss": 0.3554367125034332,
      "eval_runtime": 5.9138,
      "eval_samples_per_second": 634.105,
      "eval_steps_per_second": 9.977,
      "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
}