trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.888268156424581,
  "eval_steps": 500,
  "global_step": 3500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7,
      "grad_norm": 6.08462381362915,
      "learning_rate": 1.7206703910614527e-05,
      "loss": 0.3865,
      "step": 500
    },
    {
      "epoch": 1.4,
      "grad_norm": 5.282430648803711,
      "learning_rate": 1.4413407821229052e-05,
      "loss": 0.2899,
      "step": 1000
    },
    {
      "epoch": 2.09,
      "grad_norm": 4.065394878387451,
      "learning_rate": 1.1620111731843577e-05,
      "loss": 0.2214,
      "step": 1500
    },
    {
      "epoch": 2.79,
      "grad_norm": 8.418371200561523,
      "learning_rate": 8.826815642458101e-06,
      "loss": 0.1093,
      "step": 2000
    },
    {
      "epoch": 3.49,
      "grad_norm": 20.9011173248291,
      "learning_rate": 6.033519553072626e-06,
      "loss": 0.0653,
      "step": 2500
    },
    {
      "epoch": 4.19,
      "grad_norm": 4.803398132324219,
      "learning_rate": 3.240223463687151e-06,
      "loss": 0.0406,
      "step": 3000
    },
    {
      "epoch": 4.89,
      "grad_norm": 0.0067610470578074455,
      "learning_rate": 4.46927374301676e-07,
      "loss": 0.0164,
      "step": 3500
    }
  ],
  "logging_steps": 500,
  "max_steps": 3580,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3179834312329860.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}