trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "grad_norm": 5.483969982060544,
      "learning_rate": 6.666666666666667e-05,
      "loss": 3.748,
      "step": 1
    },
    {
      "epoch": 1.0,
      "grad_norm": 7.090502665836568,
      "learning_rate": 0.00019594929736144976,
      "loss": 3.7295,
      "step": 5
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.4484840015141356,
      "learning_rate": 0.00015406408174555976,
      "loss": 3.5598,
      "step": 10
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5836432105312597,
      "learning_rate": 8.57685161726715e-05,
      "loss": 3.4156,
      "step": 15
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.5347375258903977,
      "learning_rate": 2.4425042564574184e-05,
      "loss": 3.3484,
      "step": 20
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.2120448685091212,
      "learning_rate": 0.0,
      "loss": 3.3277,
      "step": 25
    },
    {
      "epoch": 5.0,
      "step": 25,
      "total_flos": 1192860057600.0,
      "train_loss": 3.476953125,
      "train_runtime": 86.482,
      "train_samples_per_second": 16.593,
      "train_steps_per_second": 0.289
    }
  ],
  "logging_steps": 5,
  "max_steps": 25,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 100,
  "total_flos": 1192860057600.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}