checkpoint-15/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.67,
      "learning_rate": 0.0002,
      "loss": 1.111,
      "step": 2
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00016923076923076923,
      "loss": 1.0667,
      "step": 4
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.9187,
      "step": 6
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.842,
      "step": 8
    },
    {
      "epoch": 3.33,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.8033,
      "step": 10
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.7776,
      "step": 12
    },
    {
      "epoch": 4.67,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.7449,
      "step": 14
    }
  ],
  "logging_steps": 2,
  "max_steps": 15,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 2490332432302080.0,
  "trial_name": null,
  "trial_params": null
}