trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.109542245502164,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.73,
      "eval_loss": 0.04153425619006157,
      "eval_runtime": 134.1421,
      "eval_samples_per_second": 18.615,
      "eval_steps_per_second": 1.17,
      "step": 200
    },
    {
      "epoch": 1.46,
      "eval_loss": 0.03776169568300247,
      "eval_runtime": 133.8056,
      "eval_samples_per_second": 18.661,
      "eval_steps_per_second": 1.173,
      "step": 400
    },
    {
      "epoch": 1.82,
      "learning_rate": 8.60545524394929e-05,
      "loss": 1.1974,
      "step": 500
    },
    {
      "epoch": 2.19,
      "eval_loss": 0.038170840591192245,
      "eval_runtime": 133.7976,
      "eval_samples_per_second": 18.663,
      "eval_steps_per_second": 1.173,
      "step": 600
    },
    {
      "epoch": 2.92,
      "eval_loss": 0.03970686346292496,
      "eval_runtime": 133.7983,
      "eval_samples_per_second": 18.662,
      "eval_steps_per_second": 1.173,
      "step": 800
    },
    {
      "epoch": 3.64,
      "learning_rate": 6.684594698424894e-05,
      "loss": 0.0176,
      "step": 1000
    },
    {
      "epoch": 3.64,
      "eval_loss": 0.040207117795944214,
      "eval_runtime": 133.8022,
      "eval_samples_per_second": 18.662,
      "eval_steps_per_second": 1.173,
      "step": 1000
    },
    {
      "epoch": 4.37,
      "eval_loss": 0.04497513547539711,
      "eval_runtime": 133.8142,
      "eval_samples_per_second": 18.66,
      "eval_steps_per_second": 1.173,
      "step": 1200
    },
    {
      "epoch": 5.1,
      "eval_loss": 0.04816456884145737,
      "eval_runtime": 133.7956,
      "eval_samples_per_second": 18.663,
      "eval_steps_per_second": 1.173,
      "step": 1400
    },
    {
      "epoch": 5.47,
      "learning_rate": 4.7637341529005e-05,
      "loss": 0.0096,
      "step": 1500
    },
    {
      "epoch": 5.83,
      "eval_loss": 0.04493580386042595,
      "eval_runtime": 133.8039,
      "eval_samples_per_second": 18.662,
      "eval_steps_per_second": 1.173,
      "step": 1600
    },
    {
      "epoch": 6.56,
      "eval_loss": 0.04887638986110687,
      "eval_runtime": 133.8071,
      "eval_samples_per_second": 18.661,
      "eval_steps_per_second": 1.173,
      "step": 1800
    },
    {
      "epoch": 7.29,
      "learning_rate": 2.8428736073761048e-05,
      "loss": 0.0049,
      "step": 2000
    },
    {
      "epoch": 7.29,
      "eval_loss": 0.05439848080277443,
      "eval_runtime": 133.7974,
      "eval_samples_per_second": 18.663,
      "eval_steps_per_second": 1.173,
      "step": 2000
    },
    {
      "epoch": 8.02,
      "eval_loss": 0.055828921496868134,
      "eval_runtime": 133.8004,
      "eval_samples_per_second": 18.662,
      "eval_steps_per_second": 1.173,
      "step": 2200
    },
    {
      "epoch": 8.75,
      "eval_loss": 0.056505098938941956,
      "eval_runtime": 133.7919,
      "eval_samples_per_second": 18.663,
      "eval_steps_per_second": 1.173,
      "step": 2400
    },
    {
      "epoch": 9.11,
      "learning_rate": 9.220130618517096e-06,
      "loss": 0.0028,
      "step": 2500
    }
  ],
  "max_steps": 2740,
  "num_train_epochs": 10,
  "total_flos": 1.9485556373716992e+17,
  "trial_name": null,
  "trial_params": null
}