trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 18.51851851851852,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.85,
      "learning_rate": 5.401e-05,
      "loss": 2.5377,
      "step": 100
    },
    {
      "epoch": 3.7,
      "learning_rate": 4.8020000000000004e-05,
      "loss": 1.7685,
      "step": 200
    },
    {
      "epoch": 5.56,
      "learning_rate": 4.203e-05,
      "loss": 1.4614,
      "step": 300
    },
    {
      "epoch": 7.41,
      "learning_rate": 3.604e-05,
      "loss": 1.282,
      "step": 400
    },
    {
      "epoch": 9.26,
      "learning_rate": 3.0050000000000002e-05,
      "loss": 1.157,
      "step": 500
    },
    {
      "epoch": 11.11,
      "learning_rate": 2.406e-05,
      "loss": 1.0986,
      "step": 600
    },
    {
      "epoch": 12.96,
      "learning_rate": 1.807e-05,
      "loss": 1.0388,
      "step": 700
    },
    {
      "epoch": 14.81,
      "learning_rate": 1.2079999999999998e-05,
      "loss": 0.9946,
      "step": 800
    },
    {
      "epoch": 16.67,
      "learning_rate": 6.0899999999999984e-06,
      "loss": 0.9791,
      "step": 900
    },
    {
      "epoch": 18.52,
      "learning_rate": 1e-07,
      "loss": 0.9462,
      "step": 1000
    },
    {
      "epoch": 18.52,
      "step": 1000,
      "total_flos": 2.7666045298173542e+17,
      "train_loss": 1.3263815078735353,
      "train_runtime": 482.7688,
      "train_samples_per_second": 33.142,
      "train_steps_per_second": 2.071
    }
  ],
  "max_steps": 1000,
  "num_train_epochs": 19,
  "total_flos": 2.7666045298173542e+17,
  "trial_name": null,
  "trial_params": null
}