node/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 39.02439024390244,
  "global_step": 1600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.44,
      "learning_rate": 3.7600000000000004e-06,
      "loss": 3.0404,
      "step": 100
    },
    {
      "epoch": 2.44,
      "eval_loss": 2.7579736709594727,
      "eval_runtime": 4.5474,
      "eval_samples_per_second": 4.618,
      "eval_steps_per_second": 0.66,
      "step": 100
    },
    {
      "epoch": 4.88,
      "learning_rate": 7.76e-06,
      "loss": 2.3488,
      "step": 200
    },
    {
      "epoch": 4.88,
      "eval_loss": 2.4349780082702637,
      "eval_runtime": 4.0455,
      "eval_samples_per_second": 5.191,
      "eval_steps_per_second": 0.742,
      "step": 200
    },
    {
      "epoch": 7.32,
      "learning_rate": 1.172e-05,
      "loss": 1.9928,
      "step": 300
    },
    {
      "epoch": 7.32,
      "eval_loss": 2.113464832305908,
      "eval_runtime": 4.0677,
      "eval_samples_per_second": 5.163,
      "eval_steps_per_second": 0.738,
      "step": 300
    },
    {
      "epoch": 9.76,
      "learning_rate": 1.5720000000000002e-05,
      "loss": 1.7819,
      "step": 400
    },
    {
      "epoch": 9.76,
      "eval_loss": 1.968690276145935,
      "eval_runtime": 4.0436,
      "eval_samples_per_second": 5.193,
      "eval_steps_per_second": 0.742,
      "step": 400
    },
    {
      "epoch": 12.2,
      "learning_rate": 1.972e-05,
      "loss": 1.5877,
      "step": 500
    },
    {
      "epoch": 12.2,
      "eval_loss": 1.7520062923431396,
      "eval_runtime": 4.029,
      "eval_samples_per_second": 5.212,
      "eval_steps_per_second": 0.745,
      "step": 500
    },
    {
      "epoch": 14.63,
      "learning_rate": 1.836842105263158e-05,
      "loss": 1.4275,
      "step": 600
    },
    {
      "epoch": 14.63,
      "eval_loss": 1.7901742458343506,
      "eval_runtime": 4.0456,
      "eval_samples_per_second": 5.191,
      "eval_steps_per_second": 0.742,
      "step": 600
    },
    {
      "epoch": 17.07,
      "learning_rate": 1.66140350877193e-05,
      "loss": 1.1999,
      "step": 700
    },
    {
      "epoch": 17.07,
      "eval_loss": 1.8930068016052246,
      "eval_runtime": 4.0412,
      "eval_samples_per_second": 5.196,
      "eval_steps_per_second": 0.742,
      "step": 700
    },
    {
      "epoch": 19.51,
      "learning_rate": 1.4859649122807018e-05,
      "loss": 1.2049,
      "step": 800
    },
    {
      "epoch": 19.51,
      "eval_loss": 1.6747853755950928,
      "eval_runtime": 4.0641,
      "eval_samples_per_second": 5.167,
      "eval_steps_per_second": 0.738,
      "step": 800
    },
    {
      "epoch": 21.95,
      "learning_rate": 1.310526315789474e-05,
      "loss": 1.0865,
      "step": 900
    },
    {
      "epoch": 21.95,
      "eval_loss": 1.642054796218872,
      "eval_runtime": 4.0327,
      "eval_samples_per_second": 5.207,
      "eval_steps_per_second": 0.744,
      "step": 900
    },
    {
      "epoch": 24.39,
      "learning_rate": 1.136842105263158e-05,
      "loss": 1.0559,
      "step": 1000
    },
    {
      "epoch": 24.39,
      "eval_loss": 1.6494040489196777,
      "eval_runtime": 4.1243,
      "eval_samples_per_second": 5.092,
      "eval_steps_per_second": 0.727,
      "step": 1000
    },
    {
      "epoch": 26.83,
      "learning_rate": 9.614035087719298e-06,
      "loss": 0.9572,
      "step": 1100
    },
    {
      "epoch": 26.83,
      "eval_loss": 1.6545538902282715,
      "eval_runtime": 4.1781,
      "eval_samples_per_second": 5.026,
      "eval_steps_per_second": 0.718,
      "step": 1100
    },
    {
      "epoch": 29.27,
      "learning_rate": 7.859649122807018e-06,
      "loss": 0.91,
      "step": 1200
    },
    {
      "epoch": 29.27,
      "eval_loss": 1.637965440750122,
      "eval_runtime": 4.203,
      "eval_samples_per_second": 4.996,
      "eval_steps_per_second": 0.714,
      "step": 1200
    },
    {
      "epoch": 31.71,
      "learning_rate": 6.105263157894738e-06,
      "loss": 0.8774,
      "step": 1300
    },
    {
      "epoch": 31.71,
      "eval_loss": 1.5634162425994873,
      "eval_runtime": 4.1567,
      "eval_samples_per_second": 5.052,
      "eval_steps_per_second": 0.722,
      "step": 1300
    },
    {
      "epoch": 34.15,
      "learning_rate": 4.350877192982457e-06,
      "loss": 0.8501,
      "step": 1400
    },
    {
      "epoch": 34.15,
      "eval_loss": 1.613369345664978,
      "eval_runtime": 4.1441,
      "eval_samples_per_second": 5.067,
      "eval_steps_per_second": 0.724,
      "step": 1400
    },
    {
      "epoch": 36.59,
      "learning_rate": 2.5964912280701754e-06,
      "loss": 0.8121,
      "step": 1500
    },
    {
      "epoch": 36.59,
      "eval_loss": 1.5846848487854004,
      "eval_runtime": 4.1418,
      "eval_samples_per_second": 5.07,
      "eval_steps_per_second": 0.724,
      "step": 1500
    },
    {
      "epoch": 39.02,
      "learning_rate": 8.421052631578948e-07,
      "loss": 0.8692,
      "step": 1600
    },
    {
      "epoch": 39.02,
      "eval_loss": 1.591667890548706,
      "eval_runtime": 4.1152,
      "eval_samples_per_second": 5.103,
      "eval_steps_per_second": 0.729,
      "step": 1600
    }
  ],
  "max_steps": 1640,
  "num_train_epochs": 40,
  "total_flos": 3.8572914118754304e+18,
  "trial_name": null,
  "trial_params": null
}