checkpoint-3083/trainer_state.json

{
  "best_metric": 0.34020188450813293,
  "best_model_checkpoint": "attractive_faces_celebs_detection/checkpoint-3083",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 3083,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "grad_norm": 1.2501987218856812,
      "learning_rate": 2.5548961424332345e-07,
      "loss": 0.3544,
      "step": 500
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.8902180194854736,
      "learning_rate": 2.060336300692384e-07,
      "loss": 0.3567,
      "step": 1000
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.0417747497558594,
      "learning_rate": 1.565776458951533e-07,
      "loss": 0.3574,
      "step": 1500
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.3442531824111938,
      "learning_rate": 1.0712166172106825e-07,
      "loss": 0.3558,
      "step": 2000
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.7245718240737915,
      "learning_rate": 5.766567754698319e-08,
      "loss": 0.3562,
      "step": 2500
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.2036902904510498,
      "learning_rate": 8.20969337289812e-09,
      "loss": 0.3545,
      "step": 3000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8378274268104776,
      "eval_loss": 0.34020188450813293,
      "eval_runtime": 95.2822,
      "eval_samples_per_second": 108.982,
      "eval_steps_per_second": 3.411,
      "step": 3083
    }
  ],
  "logging_steps": 500,
  "max_steps": 3083,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 1.5287774695145939e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}