checkpoint-912/trainer_state.json

{
  "best_metric": 0.25590604543685913,
  "best_model_checkpoint": "faces_age_detection/checkpoint-912",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 912,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9012954966070327,
      "eval_loss": 0.26023536920547485,
      "eval_runtime": 28.2604,
      "eval_samples_per_second": 114.719,
      "eval_steps_per_second": 3.609,
      "step": 456
    },
    {
      "epoch": 1.1,
      "grad_norm": 6.117819786071777,
      "learning_rate": 4.779582366589327e-07,
      "loss": 0.1137,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9031462060456509,
      "eval_loss": 0.25590604543685913,
      "eval_runtime": 28.5203,
      "eval_samples_per_second": 113.673,
      "eval_steps_per_second": 3.576,
      "step": 912
    }
  ],
  "logging_steps": 500,
  "max_steps": 912,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 4.5209231933567386e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}