checkpoint-21000/trainer_state.json

{
  "best_metric": 0.03351227194070816,
  "best_model_checkpoint": "/content/drive/MyDrive/Doutorado/Experimentos/colabs/trainer/checkpoint-21000",
  "epoch": 7.18439958946288,
  "global_step": 21000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.03,
      "learning_rate": 1.91811152925077e-05,
      "loss": 0.1619,
      "step": 3000
    },
    {
      "epoch": 1.03,
      "eval_loss": 0.05028245225548744,
      "eval_runtime": 27.6275,
      "eval_samples_per_second": 33.445,
      "eval_steps_per_second": 4.199,
      "step": 3000
    },
    {
      "epoch": 2.05,
      "learning_rate": 1.836004105371194e-05,
      "loss": 0.0548,
      "step": 6000
    },
    {
      "epoch": 2.05,
      "eval_loss": 0.04163406416773796,
      "eval_runtime": 27.2823,
      "eval_samples_per_second": 33.868,
      "eval_steps_per_second": 4.252,
      "step": 6000
    },
    {
      "epoch": 3.08,
      "learning_rate": 1.753896681491618e-05,
      "loss": 0.0441,
      "step": 9000
    },
    {
      "epoch": 3.08,
      "eval_loss": 0.037465475499629974,
      "eval_runtime": 26.8832,
      "eval_samples_per_second": 34.371,
      "eval_steps_per_second": 4.315,
      "step": 9000
    },
    {
      "epoch": 4.11,
      "learning_rate": 1.6717892576120426e-05,
      "loss": 0.0391,
      "step": 12000
    },
    {
      "epoch": 4.11,
      "eval_loss": 0.03538713976740837,
      "eval_runtime": 28.4361,
      "eval_samples_per_second": 32.494,
      "eval_steps_per_second": 4.079,
      "step": 12000
    },
    {
      "epoch": 5.13,
      "learning_rate": 1.589681833732467e-05,
      "loss": 0.0335,
      "step": 15000
    },
    {
      "epoch": 5.13,
      "eval_loss": 0.03461988642811775,
      "eval_runtime": 28.0681,
      "eval_samples_per_second": 32.92,
      "eval_steps_per_second": 4.133,
      "step": 15000
    },
    {
      "epoch": 6.16,
      "learning_rate": 1.5075744098528911e-05,
      "loss": 0.0304,
      "step": 18000
    },
    {
      "epoch": 6.16,
      "eval_loss": 0.033533450216054916,
      "eval_runtime": 27.7618,
      "eval_samples_per_second": 33.283,
      "eval_steps_per_second": 4.178,
      "step": 18000
    },
    {
      "epoch": 7.18,
      "learning_rate": 1.4254943551146082e-05,
      "loss": 0.0273,
      "step": 21000
    },
    {
      "epoch": 7.18,
      "eval_loss": 0.03351227194070816,
      "eval_runtime": 27.8867,
      "eval_samples_per_second": 33.134,
      "eval_steps_per_second": 4.16,
      "step": 21000
    }
  ],
  "max_steps": 73075,
  "num_train_epochs": 25,
  "total_flos": 3.523012168812466e+19,
  "trial_name": null,
  "trial_params": null
}