trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "global_step": 7550,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 1.664459161147903e-05,
      "loss": 12.0516,
      "step": 377
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.9978970892535284e-05,
      "loss": 2.6353,
      "step": 755
    },
    {
      "epoch": 3.0,
      "learning_rate": 1.9893847819221357e-05,
      "loss": 2.1592,
      "step": 1132
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.974340799082944e-05,
      "loss": 1.9294,
      "step": 1510
    },
    {
      "epoch": 5.0,
      "learning_rate": 1.9529432771098294e-05,
      "loss": 1.8008,
      "step": 1887
    },
    {
      "epoch": 6.0,
      "learning_rate": 1.9252197244209404e-05,
      "loss": 1.7041,
      "step": 2265
    },
    {
      "epoch": 7.0,
      "learning_rate": 1.8914984166597534e-05,
      "loss": 1.5976,
      "step": 2642
    },
    {
      "epoch": 8.0,
      "learning_rate": 1.851822705567445e-05,
      "loss": 1.5166,
      "step": 3020
    },
    {
      "epoch": 9.0,
      "learning_rate": 1.8066623927504456e-05,
      "loss": 1.5306,
      "step": 3397
    },
    {
      "epoch": 10.0,
      "learning_rate": 1.756075535808614e-05,
      "loss": 1.6441,
      "step": 3775
    },
    {
      "epoch": 11.0,
      "learning_rate": 1.7006611356420277e-05,
      "loss": 1.4489,
      "step": 4152
    },
    {
      "epoch": 12.0,
      "learning_rate": 1.6404904323921e-05,
      "loss": 1.3043,
      "step": 4530
    },
    {
      "epoch": 13.0,
      "learning_rate": 1.576275909648123e-05,
      "loss": 1.2039,
      "step": 4907
    },
    {
      "epoch": 14.0,
      "learning_rate": 1.508100120891996e-05,
      "loss": 1.1221,
      "step": 5285
    },
    {
      "epoch": 15.0,
      "learning_rate": 1.436770338235432e-05,
      "loss": 1.0508,
      "step": 5662
    },
    {
      "epoch": 16.0,
      "learning_rate": 1.3623782624599628e-05,
      "loss": 0.9998,
      "step": 6040
    },
    {
      "epoch": 17.0,
      "learning_rate": 1.2858047729686983e-05,
      "loss": 0.9482,
      "step": 6417
    },
    {
      "epoch": 18.0,
      "learning_rate": 1.2071483117948186e-05,
      "loss": 0.8819,
      "step": 6795
    },
    {
      "epoch": 19.0,
      "learning_rate": 1.1273402530913804e-05,
      "loss": 0.8404,
      "step": 7172
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.0464831972180532e-05,
      "loss": 0.7952,
      "step": 7550
    }
  ],
  "max_steps": 15080,
  "num_train_epochs": 40,
  "total_flos": 5193491919405056.0,
  "trial_name": null,
  "trial_params": null
}