checkpoint-1312/trainer_state.json

{
  "best_metric": 0.48922964930534363,
  "best_model_checkpoint": "marvel_heroes_image_detection/checkpoint-1312",
  "epoch": 16.0,
  "eval_steps": 500,
  "global_step": 1312,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8984615384615384,
      "eval_loss": 0.49428293108940125,
      "eval_runtime": 12.0605,
      "eval_samples_per_second": 53.895,
      "eval_steps_per_second": 6.799,
      "step": 82
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8969230769230769,
      "eval_loss": 0.49431312084198,
      "eval_runtime": 12.3803,
      "eval_samples_per_second": 52.503,
      "eval_steps_per_second": 6.623,
      "step": 164
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9015384615384615,
      "eval_loss": 0.49280208349227905,
      "eval_runtime": 12.4968,
      "eval_samples_per_second": 52.013,
      "eval_steps_per_second": 6.562,
      "step": 246
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8969230769230769,
      "eval_loss": 0.49426719546318054,
      "eval_runtime": 12.057,
      "eval_samples_per_second": 53.91,
      "eval_steps_per_second": 6.801,
      "step": 328
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8984615384615384,
      "eval_loss": 0.4925532042980194,
      "eval_runtime": 12.0769,
      "eval_samples_per_second": 53.822,
      "eval_steps_per_second": 6.79,
      "step": 410
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8938461538461538,
      "eval_loss": 0.49220046401023865,
      "eval_runtime": 12.1699,
      "eval_samples_per_second": 53.411,
      "eval_steps_per_second": 6.738,
      "step": 492
    },
    {
      "epoch": 6.1,
      "grad_norm": 4.925354480743408,
      "learning_rate": 7.169811320754716e-07,
      "loss": 0.5614,
      "step": 500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8876923076923077,
      "eval_loss": 0.49260208010673523,
      "eval_runtime": 13.062,
      "eval_samples_per_second": 49.763,
      "eval_steps_per_second": 6.278,
      "step": 574
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8984615384615384,
      "eval_loss": 0.4912846088409424,
      "eval_runtime": 12.3802,
      "eval_samples_per_second": 52.503,
      "eval_steps_per_second": 6.623,
      "step": 656
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8923076923076924,
      "eval_loss": 0.4908252954483032,
      "eval_runtime": 12.0634,
      "eval_samples_per_second": 53.882,
      "eval_steps_per_second": 6.797,
      "step": 738
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8923076923076924,
      "eval_loss": 0.4904528260231018,
      "eval_runtime": 12.2539,
      "eval_samples_per_second": 53.044,
      "eval_steps_per_second": 6.692,
      "step": 820
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.8892307692307693,
      "eval_loss": 0.4904911518096924,
      "eval_runtime": 12.1087,
      "eval_samples_per_second": 53.68,
      "eval_steps_per_second": 6.772,
      "step": 902
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.8938461538461538,
      "eval_loss": 0.48976370692253113,
      "eval_runtime": 12.1182,
      "eval_samples_per_second": 53.639,
      "eval_steps_per_second": 6.767,
      "step": 984
    },
    {
      "epoch": 12.2,
      "grad_norm": 4.339879512786865,
      "learning_rate": 4.0251572327044023e-07,
      "loss": 0.519,
      "step": 1000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.8892307692307693,
      "eval_loss": 0.489905446767807,
      "eval_runtime": 12.1611,
      "eval_samples_per_second": 53.449,
      "eval_steps_per_second": 6.743,
      "step": 1066
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.8923076923076924,
      "eval_loss": 0.48949161171913147,
      "eval_runtime": 12.1412,
      "eval_samples_per_second": 53.537,
      "eval_steps_per_second": 6.754,
      "step": 1148
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.8923076923076924,
      "eval_loss": 0.48962101340293884,
      "eval_runtime": 12.0964,
      "eval_samples_per_second": 53.735,
      "eval_steps_per_second": 6.779,
      "step": 1230
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8907692307692308,
      "eval_loss": 0.48922964930534363,
      "eval_runtime": 11.9519,
      "eval_samples_per_second": 54.385,
      "eval_steps_per_second": 6.861,
      "step": 1312
    }
  ],
  "logging_steps": 500,
  "max_steps": 1640,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "total_flos": 3.2213602471819346e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}