trainer_state.json

{
  "best_metric": 0.42106666666666664,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0611/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": Infinity,
      "learning_rate": 0.0002911677661332091,
      "loss": 3.3764,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.21706666666666666,
      "eval_loss": 2.9657974243164062,
      "eval_runtime": 10.3619,
      "eval_samples_per_second": 361.902,
      "eval_steps_per_second": 5.694,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 42.547569274902344,
      "learning_rate": 0.0002653099348375843,
      "loss": 2.8185,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2544,
      "eval_loss": 2.7827277183532715,
      "eval_runtime": 10.15,
      "eval_samples_per_second": 369.46,
      "eval_steps_per_second": 5.813,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 27.80939483642578,
      "learning_rate": 0.00022554402360715464,
      "loss": 2.5593,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2944,
      "eval_loss": 2.606044292449951,
      "eval_runtime": 10.5521,
      "eval_samples_per_second": 355.381,
      "eval_steps_per_second": 5.591,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 28.772682189941406,
      "learning_rate": 0.00017666638821865508,
      "loss": 2.3162,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.33866666666666667,
      "eval_loss": 2.395925998687744,
      "eval_runtime": 10.2569,
      "eval_samples_per_second": 365.609,
      "eval_steps_per_second": 5.752,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.453884124755859,
      "learning_rate": 0.00012457239285701023,
      "loss": 2.0567,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3741333333333333,
      "eval_loss": 2.3038296699523926,
      "eval_runtime": 10.5622,
      "eval_samples_per_second": 355.041,
      "eval_steps_per_second": 5.586,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 39.3098030090332,
      "learning_rate": 7.554534218832392e-05,
      "loss": 1.7445,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.40426666666666666,
      "eval_loss": 2.21852707862854,
      "eval_runtime": 10.4023,
      "eval_samples_per_second": 360.496,
      "eval_steps_per_second": 5.672,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 44.18996810913086,
      "learning_rate": 3.5498622085445925e-05,
      "loss": 1.3158,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4184,
      "eval_loss": 2.2666943073272705,
      "eval_runtime": 10.4957,
      "eval_samples_per_second": 357.289,
      "eval_steps_per_second": 5.621,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": Infinity,
      "learning_rate": 9.31693256552663e-06,
      "loss": 0.8358,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.42106666666666664,
      "eval_loss": 2.698033094406128,
      "eval_runtime": 10.2947,
      "eval_samples_per_second": 364.266,
      "eval_steps_per_second": 5.731,
      "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
}