checkpoint-10/trainer_state.json

{
  "best_metric": 9.324381828308105,
  "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/lm-training/src/outputs/2023-11-15/09-23-00/checkpoints/checkpoint-10",
  "epoch": 0.08,
  "eval_steps": 10,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 4.866666666666667e-05,
      "loss": 9.9717,
      "step": 10
    },
    {
      "epoch": 0.08,
      "eval_loss": 9.324381828308105,
      "eval_runtime": 2.0042,
      "eval_samples_per_second": 498.949,
      "eval_steps_per_second": 62.369,
      "step": 10
    }
  ],
  "logging_steps": 10,
  "max_steps": 375,
  "num_train_epochs": 3,
  "save_steps": 10,
  "total_flos": 1184920289280.0,
  "trial_name": null,
  "trial_params": null
}