trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 9237,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 4.729349355851467e-05,
      "loss": 1.9325,
      "step": 500
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.458698711702934e-05,
      "loss": 1.7345,
      "step": 1000
    },
    {
      "epoch": 0.49,
      "learning_rate": 4.188048067554401e-05,
      "loss": 1.6281,
      "step": 1500
    },
    {
      "epoch": 0.65,
      "learning_rate": 3.917397423405868e-05,
      "loss": 1.5893,
      "step": 2000
    },
    {
      "epoch": 0.81,
      "learning_rate": 3.646746779257334e-05,
      "loss": 1.5438,
      "step": 2500
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.376096135108802e-05,
      "loss": 1.5096,
      "step": 3000
    },
    {
      "epoch": 1.14,
      "learning_rate": 3.105445490960269e-05,
      "loss": 1.2386,
      "step": 3500
    },
    {
      "epoch": 1.3,
      "learning_rate": 2.8347948468117357e-05,
      "loss": 1.1895,
      "step": 4000
    },
    {
      "epoch": 1.46,
      "learning_rate": 2.5641442026632023e-05,
      "loss": 1.1553,
      "step": 4500
    },
    {
      "epoch": 1.62,
      "learning_rate": 2.2934935585146693e-05,
      "loss": 1.1808,
      "step": 5000
    },
    {
      "epoch": 1.79,
      "learning_rate": 2.0228429143661362e-05,
      "loss": 1.1716,
      "step": 5500
    },
    {
      "epoch": 1.95,
      "learning_rate": 1.752192270217603e-05,
      "loss": 1.157,
      "step": 6000
    },
    {
      "epoch": 2.11,
      "learning_rate": 1.4815416260690701e-05,
      "loss": 1.0016,
      "step": 6500
    },
    {
      "epoch": 2.27,
      "learning_rate": 1.210890981920537e-05,
      "loss": 0.9479,
      "step": 7000
    },
    {
      "epoch": 2.44,
      "learning_rate": 9.40240337772004e-06,
      "loss": 0.9462,
      "step": 7500
    },
    {
      "epoch": 2.6,
      "learning_rate": 6.695896936234708e-06,
      "loss": 0.9277,
      "step": 8000
    },
    {
      "epoch": 2.76,
      "learning_rate": 3.9893904947493776e-06,
      "loss": 0.9245,
      "step": 8500
    },
    {
      "epoch": 2.92,
      "learning_rate": 1.2828840532640467e-06,
      "loss": 0.9232,
      "step": 9000
    },
    {
      "epoch": 3.0,
      "step": 9237,
      "total_flos": 1.2276424759345152e+16,
      "train_loss": 1.2522439968021992,
      "train_runtime": 2736.5651,
      "train_samples_per_second": 33.745,
      "train_steps_per_second": 3.375
    }
  ],
  "max_steps": 9237,
  "num_train_epochs": 3,
  "total_flos": 1.2276424759345152e+16,
  "trial_name": null,
  "trial_params": null
}