last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 50.0,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.5,
      "learning_rate": 0.0002,
      "loss": 1.3126,
      "step": 10
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0002,
      "loss": 0.4634,
      "step": 20
    },
    {
      "epoch": 7.5,
      "learning_rate": 0.0002,
      "loss": 0.18,
      "step": 30
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.0002,
      "loss": 0.0961,
      "step": 40
    },
    {
      "epoch": 12.5,
      "learning_rate": 0.0002,
      "loss": 0.0536,
      "step": 50
    },
    {
      "epoch": 15.0,
      "learning_rate": 0.0002,
      "loss": 0.0325,
      "step": 60
    },
    {
      "epoch": 17.5,
      "learning_rate": 0.0002,
      "loss": 0.0207,
      "step": 70
    },
    {
      "epoch": 20.0,
      "learning_rate": 0.0002,
      "loss": 0.0161,
      "step": 80
    },
    {
      "epoch": 22.5,
      "learning_rate": 0.0002,
      "loss": 0.0122,
      "step": 90
    },
    {
      "epoch": 25.0,
      "learning_rate": 0.0002,
      "loss": 0.0113,
      "step": 100
    },
    {
      "epoch": 27.5,
      "learning_rate": 0.0002,
      "loss": 0.0103,
      "step": 110
    },
    {
      "epoch": 30.0,
      "learning_rate": 0.0002,
      "loss": 0.0102,
      "step": 120
    },
    {
      "epoch": 32.5,
      "learning_rate": 0.0002,
      "loss": 0.0095,
      "step": 130
    },
    {
      "epoch": 35.0,
      "learning_rate": 0.0002,
      "loss": 0.0094,
      "step": 140
    },
    {
      "epoch": 37.5,
      "learning_rate": 0.0002,
      "loss": 0.0091,
      "step": 150
    },
    {
      "epoch": 40.0,
      "learning_rate": 0.0002,
      "loss": 0.009,
      "step": 160
    },
    {
      "epoch": 42.5,
      "learning_rate": 0.0002,
      "loss": 0.0088,
      "step": 170
    },
    {
      "epoch": 45.0,
      "learning_rate": 0.0002,
      "loss": 0.0089,
      "step": 180
    },
    {
      "epoch": 47.5,
      "learning_rate": 0.0002,
      "loss": 0.0088,
      "step": 190
    },
    {
      "epoch": 50.0,
      "learning_rate": 0.0002,
      "loss": 0.009,
      "step": 200
    }
  ],
  "max_steps": 200,
  "num_train_epochs": 50,
  "total_flos": 6.5384196046848e+16,
  "trial_name": null,
  "trial_params": null
}