checkpoint-540/trainer_state.json

{
  "best_metric": 0.31217730045318604,
  "best_model_checkpoint": "outputs/t5-base-detox/checkpoint-540",
  "epoch": 3.9953917050691246,
  "global_step": 540,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 9e-05,
      "loss": 1.337,
      "step": 135
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.48098286986351013,
      "eval_runtime": 18.2812,
      "eval_samples_per_second": 169.573,
      "eval_steps_per_second": 21.224,
      "step": 135
    },
    {
      "epoch": 2.0,
      "learning_rate": 8e-05,
      "loss": 0.5238,
      "step": 270
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.38857772946357727,
      "eval_runtime": 18.7764,
      "eval_samples_per_second": 165.101,
      "eval_steps_per_second": 20.664,
      "step": 270
    },
    {
      "epoch": 3.0,
      "learning_rate": 7e-05,
      "loss": 0.4301,
      "step": 405
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.33775028586387634,
      "eval_runtime": 18.505,
      "eval_samples_per_second": 167.522,
      "eval_steps_per_second": 20.967,
      "step": 405
    },
    {
      "epoch": 4.0,
      "learning_rate": 6e-05,
      "loss": 0.3755,
      "step": 540
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.31217730045318604,
      "eval_runtime": 18.8887,
      "eval_samples_per_second": 164.119,
      "eval_steps_per_second": 20.541,
      "step": 540
    }
  ],
  "max_steps": 1350,
  "num_train_epochs": 10,
  "total_flos": 4680745310822400.0,
  "trial_name": null,
  "trial_params": null
}