trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 2950,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "learning_rate": 8.305084745762712e-07,
      "loss": 0.634,
      "step": 500
    },
    {
      "epoch": 0.34,
      "learning_rate": 6.610169491525423e-07,
      "loss": 0.4411,
      "step": 1000
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.915254237288136e-07,
      "loss": 0.3908,
      "step": 1500
    },
    {
      "epoch": 0.68,
      "learning_rate": 3.220338983050847e-07,
      "loss": 0.3642,
      "step": 2000
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.5254237288135593e-07,
      "loss": 0.3506,
      "step": 2500
    },
    {
      "epoch": 1.0,
      "step": 2950,
      "total_flos": 5498162060794368.0,
      "train_loss": 0.4227196282855535,
      "train_runtime": 1206.6968,
      "train_samples_per_second": 39.113,
      "train_steps_per_second": 2.445
    }
  ],
  "logging_steps": 500,
  "max_steps": 2950,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 5498162060794368.0,
  "trial_name": null,
  "trial_params": null
}