trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.095890410958904,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "learning_rate": 1.9391171993911722e-05,
      "loss": 2.5862,
      "step": 500
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8782343987823442e-05,
      "loss": 1.5245,
      "step": 1000
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8173515981735163e-05,
      "loss": 1.4047,
      "step": 1500
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.756468797564688e-05,
      "loss": 1.2872,
      "step": 2000
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.69558599695586e-05,
      "loss": 1.2613,
      "step": 2500
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.634703196347032e-05,
      "loss": 1.2014,
      "step": 3000
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.573820395738204e-05,
      "loss": 1.1608,
      "step": 3500
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5129375951293761e-05,
      "loss": 1.1201,
      "step": 4000
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4520547945205482e-05,
      "loss": 1.1287,
      "step": 4500
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.39117199391172e-05,
      "loss": 1.1157,
      "step": 5000
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.059618353843689,
      "eval_runtime": 282.3733,
      "eval_samples_per_second": 37.433,
      "eval_steps_per_second": 2.341,
      "step": 5475
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.330289193302892e-05,
      "loss": 1.0706,
      "step": 5500
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2694063926940641e-05,
      "loss": 0.8605,
      "step": 6000
    }
  ],
  "max_steps": 16425,
  "num_train_epochs": 3,
  "total_flos": 1.881317051242445e+16,
  "trial_name": null,
  "trial_params": null
}