trainer_state.json

{
  "best_metric": 0.8563636363636363,
  "best_model_checkpoint": "/content/drive/MyDrive/autoTaggingProject/ViT/General/Features/TopLength/Results/model_2024-10-16_test/checkpoint-1925",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 1925,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 1.4969431161880493,
      "learning_rate": 9.81818181818182e-06,
      "loss": 1.6534,
      "step": 100
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 2.2577221393585205,
      "learning_rate": 9.636363636363638e-06,
      "loss": 1.317,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7218181818181818,
      "eval_loss": 1.0995656251907349,
      "eval_runtime": 12.6938,
      "eval_samples_per_second": 43.328,
      "eval_steps_per_second": 1.418,
      "step": 275
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 2.7553977966308594,
      "learning_rate": 9.454545454545456e-06,
      "loss": 1.0529,
      "step": 300
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 2.5488719940185547,
      "learning_rate": 9.272727272727273e-06,
      "loss": 0.9003,
      "step": 400
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 2.936871290206909,
      "learning_rate": 9.090909090909091e-06,
      "loss": 0.7955,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7890909090909091,
      "eval_loss": 0.8251153826713562,
      "eval_runtime": 11.8903,
      "eval_samples_per_second": 46.256,
      "eval_steps_per_second": 1.514,
      "step": 550
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 3.6827473640441895,
      "learning_rate": 8.90909090909091e-06,
      "loss": 0.7386,
      "step": 600
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 2.6154849529266357,
      "learning_rate": 8.727272727272728e-06,
      "loss": 0.6758,
      "step": 700
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 6.043990135192871,
      "learning_rate": 8.545454545454546e-06,
      "loss": 0.6518,
      "step": 800
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.82,
      "eval_loss": 0.7125250101089478,
      "eval_runtime": 11.5548,
      "eval_samples_per_second": 47.599,
      "eval_steps_per_second": 1.558,
      "step": 825
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 1.676995873451233,
      "learning_rate": 8.363636363636365e-06,
      "loss": 0.6034,
      "step": 900
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 2.6665966510772705,
      "learning_rate": 8.181818181818183e-06,
      "loss": 0.5865,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.419482469558716,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.5531,
      "step": 1100
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8218181818181818,
      "eval_loss": 0.5728441476821899,
      "eval_runtime": 11.88,
      "eval_samples_per_second": 46.296,
      "eval_steps_per_second": 1.515,
      "step": 1100
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 2.151674747467041,
      "learning_rate": 7.81818181818182e-06,
      "loss": 0.5151,
      "step": 1200
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 4.103756427764893,
      "learning_rate": 7.63818181818182e-06,
      "loss": 0.5118,
      "step": 1300
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.84,
      "eval_loss": 0.631469190120697,
      "eval_runtime": 11.4374,
      "eval_samples_per_second": 48.088,
      "eval_steps_per_second": 1.574,
      "step": 1375
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 2.0776937007904053,
      "learning_rate": 7.456363636363637e-06,
      "loss": 0.4878,
      "step": 1400
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 4.111904621124268,
      "learning_rate": 7.274545454545456e-06,
      "loss": 0.4802,
      "step": 1500
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 3.490161657333374,
      "learning_rate": 7.092727272727273e-06,
      "loss": 0.4713,
      "step": 1600
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.84,
      "eval_loss": 0.4893127381801605,
      "eval_runtime": 11.1344,
      "eval_samples_per_second": 49.396,
      "eval_steps_per_second": 1.617,
      "step": 1650
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 3.1339528560638428,
      "learning_rate": 6.910909090909092e-06,
      "loss": 0.4365,
      "step": 1700
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 3.686084032058716,
      "learning_rate": 6.72909090909091e-06,
      "loss": 0.4155,
      "step": 1800
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 3.6479032039642334,
      "learning_rate": 6.5472727272727275e-06,
      "loss": 0.4251,
      "step": 1900
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8563636363636363,
      "eval_loss": 0.48278242349624634,
      "eval_runtime": 11.2502,
      "eval_samples_per_second": 48.888,
      "eval_steps_per_second": 1.6,
      "step": 1925
    }
  ],
  "logging_steps": 100,
  "max_steps": 5500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 4.773677693976576e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}