checkpoint-3400/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8756116404841617,
  "eval_steps": 500,
  "global_step": 3400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "learning_rate": 1.742467164563482e-05,
      "loss": 1.5613,
      "step": 500
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.4849343291269638e-05,
      "loss": 1.2101,
      "step": 1000
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.2274014936904457e-05,
      "loss": 1.1045,
      "step": 1500
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.698686582539275e-06,
      "loss": 1.0492,
      "step": 2000
    },
    {
      "epoch": 0.64,
      "learning_rate": 7.123358228174093e-06,
      "loss": 1.0014,
      "step": 2500
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.548029873808911e-06,
      "loss": 0.9832,
      "step": 3000
    }
  ],
  "logging_steps": 500,
  "max_steps": 3883,
  "num_train_epochs": 1,
  "save_steps": 200,
  "total_flos": 1253146280067072.0,
  "trial_name": null,
  "trial_params": null
}