trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 4794,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "learning_rate": 4.478514810179391e-05,
      "loss": 1.3792,
      "step": 500
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.9570296203587817e-05,
      "loss": 1.3043,
      "step": 1000
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.435544430538173e-05,
      "loss": 1.2641,
      "step": 1500
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.9140592407175638e-05,
      "loss": 1.2319,
      "step": 2000
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.3925740508969545e-05,
      "loss": 1.2035,
      "step": 2500
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8710888610763455e-05,
      "loss": 1.2072,
      "step": 3000
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.3496036712557364e-05,
      "loss": 1.1955,
      "step": 3500
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.281184814351273e-06,
      "loss": 1.1949,
      "step": 4000
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.0663329161451816e-06,
      "loss": 1.194,
      "step": 4500
    },
    {
      "epoch": 1.0,
      "step": 4794,
      "total_flos": 2.8476642525865574e+17,
      "train_loss": 1.2382940134009472,
      "train_runtime": 11022.1698,
      "train_samples_per_second": 3.479,
      "train_steps_per_second": 0.435
    }
  ],
  "logging_steps": 500,
  "max_steps": 4794,
  "num_train_epochs": 1,
  "save_steps": -4794,
  "total_flos": 2.8476642525865574e+17,
  "trial_name": null,
  "trial_params": null
}