trainer_state.json

{
  "best_metric": 0.8506666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0885/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.9328657388687134,
      "learning_rate": 0.0001,
      "loss": 0.9182,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8154666666666667,
      "eval_loss": 0.5983468294143677,
      "eval_runtime": 13.3111,
      "eval_samples_per_second": 281.719,
      "eval_steps_per_second": 4.432,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.4143,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8234666666666667,
      "eval_loss": 0.5746971368789673,
      "eval_runtime": 13.2975,
      "eval_samples_per_second": 282.007,
      "eval_steps_per_second": 4.437,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.2861502170562744,
      "learning_rate": 0.0001,
      "loss": 0.2826,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8248,
      "eval_loss": 0.5808320045471191,
      "eval_runtime": 13.2774,
      "eval_samples_per_second": 282.435,
      "eval_steps_per_second": 4.444,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.4047479033470154,
      "learning_rate": 0.0001,
      "loss": 0.2158,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8269333333333333,
      "eval_loss": 0.6375138163566589,
      "eval_runtime": 14.1003,
      "eval_samples_per_second": 265.952,
      "eval_steps_per_second": 4.184,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 56.214229583740234,
      "learning_rate": 0.0001,
      "loss": 0.1836,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8426666666666667,
      "eval_loss": 0.6095971465110779,
      "eval_runtime": 13.345,
      "eval_samples_per_second": 281.005,
      "eval_steps_per_second": 4.421,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.670413851737976,
      "learning_rate": 0.0001,
      "loss": 0.1598,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8338666666666666,
      "eval_loss": 0.6496953368186951,
      "eval_runtime": 13.3313,
      "eval_samples_per_second": 281.292,
      "eval_steps_per_second": 4.426,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.00031049951212480664,
      "learning_rate": 0.0001,
      "loss": 0.1477,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8506666666666667,
      "eval_loss": 0.609856367111206,
      "eval_runtime": 13.37,
      "eval_samples_per_second": 280.478,
      "eval_steps_per_second": 4.413,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}