{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 18.0,
  "global_step": 15678,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.57,
      "learning_rate": 9.68108177063401e-05,
      "loss": 2.3589,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.2804882526397705,
      "eval_runtime": 46.5333,
      "eval_samples_per_second": 63.03,
      "eval_steps_per_second": 7.887,
      "step": 871
    },
    {
      "epoch": 1.15,
      "learning_rate": 9.36216354126802e-05,
      "loss": 2.3359,
      "step": 1000
    },
    {
      "epoch": 1.72,
      "learning_rate": 9.043245311902028e-05,
      "loss": 2.2563,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.2500908374786377,
      "eval_runtime": 46.4819,
      "eval_samples_per_second": 63.1,
      "eval_steps_per_second": 7.896,
      "step": 1742
    },
    {
      "epoch": 2.3,
      "learning_rate": 8.724327082536038e-05,
      "loss": 2.2043,
      "step": 2000
    },
    {
      "epoch": 2.87,
      "learning_rate": 8.405408853170048e-05,
      "loss": 2.1936,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.2418746948242188,
      "eval_runtime": 46.4009,
      "eval_samples_per_second": 63.21,
      "eval_steps_per_second": 7.909,
      "step": 2613
    },
    {
      "epoch": 3.44,
      "learning_rate": 8.086490623804057e-05,
      "loss": 2.11,
      "step": 3000
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.230112314224243,
      "eval_runtime": 46.3837,
      "eval_samples_per_second": 63.233,
      "eval_steps_per_second": 7.912,
      "step": 3484
    },
    {
      "epoch": 4.02,
      "learning_rate": 7.767572394438066e-05,
      "loss": 2.1018,
      "step": 3500
    },
    {
      "epoch": 4.59,
      "learning_rate": 7.448654165072075e-05,
      "loss": 2.0311,
      "step": 4000
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.231987476348877,
      "eval_runtime": 46.4154,
      "eval_samples_per_second": 63.19,
      "eval_steps_per_second": 7.907,
      "step": 4355
    },
    {
      "epoch": 5.17,
      "learning_rate": 7.129735935706085e-05,
      "loss": 2.0174,
      "step": 4500
    },
    {
      "epoch": 5.74,
      "learning_rate": 6.810817706340095e-05,
      "loss": 1.969,
      "step": 5000
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.227581262588501,
      "eval_runtime": 46.0273,
      "eval_samples_per_second": 63.723,
      "eval_steps_per_second": 7.974,
      "step": 5226
    },
    {
      "epoch": 6.31,
      "learning_rate": 6.491899476974103e-05,
      "loss": 1.9427,
      "step": 5500
    },
    {
      "epoch": 6.89,
      "learning_rate": 6.172981247608114e-05,
      "loss": 1.9148,
      "step": 6000
    },
    {
      "epoch": 7.0,
      "eval_loss": 2.1621322631835938,
      "eval_runtime": 45.9847,
      "eval_samples_per_second": 63.782,
      "eval_steps_per_second": 7.981,
      "step": 6097
    },
    {
      "epoch": 7.46,
      "learning_rate": 5.854063018242123e-05,
      "loss": 1.8569,
      "step": 6500
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.1876232624053955,
      "eval_runtime": 45.9436,
      "eval_samples_per_second": 63.839,
      "eval_steps_per_second": 7.988,
      "step": 6968
    },
    {
      "epoch": 8.04,
      "learning_rate": 5.535144788876132e-05,
      "loss": 1.8523,
      "step": 7000
    },
    {
      "epoch": 8.61,
      "learning_rate": 5.216226559510142e-05,
      "loss": 1.7978,
      "step": 7500
    },
    {
      "epoch": 9.0,
      "eval_loss": 2.201099395751953,
      "eval_runtime": 45.9305,
      "eval_samples_per_second": 63.857,
      "eval_steps_per_second": 7.99,
      "step": 7839
    },
    {
      "epoch": 9.18,
      "learning_rate": 4.897308330144152e-05,
      "loss": 1.7922,
      "step": 8000
    },
    {
      "epoch": 9.76,
      "learning_rate": 4.578390100778161e-05,
      "loss": 1.7602,
      "step": 8500
    },
    {
      "epoch": 10.0,
      "eval_loss": 2.1279566287994385,
      "eval_runtime": 45.9065,
      "eval_samples_per_second": 63.891,
      "eval_steps_per_second": 7.995,
      "step": 8710
    },
    {
      "epoch": 10.33,
      "learning_rate": 4.2594718714121704e-05,
      "loss": 1.7371,
      "step": 9000
    },
    {
      "epoch": 10.91,
      "learning_rate": 3.9405536420461794e-05,
      "loss": 1.7166,
      "step": 9500
    },
    {
      "epoch": 11.0,
      "eval_loss": 2.1643788814544678,
      "eval_runtime": 45.9375,
      "eval_samples_per_second": 63.848,
      "eval_steps_per_second": 7.989,
      "step": 9581
    },
    {
      "epoch": 11.48,
      "learning_rate": 3.621635412680189e-05,
      "loss": 1.6651,
      "step": 10000
    },
    {
      "epoch": 12.0,
      "eval_loss": 2.1245827674865723,
      "eval_runtime": 45.9431,
      "eval_samples_per_second": 63.84,
      "eval_steps_per_second": 7.988,
      "step": 10452
    },
    {
      "epoch": 12.06,
      "learning_rate": 3.302717183314198e-05,
      "loss": 1.6508,
      "step": 10500
    },
    {
      "epoch": 12.63,
      "learning_rate": 2.9837989539482082e-05,
      "loss": 1.6141,
      "step": 11000
    },
    {
      "epoch": 13.0,
      "eval_loss": 2.126392126083374,
      "eval_runtime": 45.9202,
      "eval_samples_per_second": 63.872,
      "eval_steps_per_second": 7.992,
      "step": 11323
    },
    {
      "epoch": 13.2,
      "learning_rate": 2.6648807245822172e-05,
      "loss": 1.6312,
      "step": 11500
    },
    {
      "epoch": 13.78,
      "learning_rate": 2.345962495216227e-05,
      "loss": 1.5759,
      "step": 12000
    },
    {
      "epoch": 14.0,
      "eval_loss": 2.114293098449707,
      "eval_runtime": 45.9021,
      "eval_samples_per_second": 63.897,
      "eval_steps_per_second": 7.995,
      "step": 12194
    },
    {
      "epoch": 14.35,
      "learning_rate": 2.0270442658502363e-05,
      "loss": 1.572,
      "step": 12500
    },
    {
      "epoch": 14.93,
      "learning_rate": 1.7081260364842456e-05,
      "loss": 1.5478,
      "step": 13000
    },
    {
      "epoch": 15.0,
      "eval_loss": 2.098154306411743,
      "eval_runtime": 45.9315,
      "eval_samples_per_second": 63.856,
      "eval_steps_per_second": 7.99,
      "step": 13065
    },
    {
      "epoch": 15.5,
      "learning_rate": 1.3892078071182548e-05,
      "loss": 1.5311,
      "step": 13500
    },
    {
      "epoch": 16.0,
      "eval_loss": 2.099287509918213,
      "eval_runtime": 45.9245,
      "eval_samples_per_second": 63.866,
      "eval_steps_per_second": 7.991,
      "step": 13936
    },
    {
      "epoch": 16.07,
      "learning_rate": 1.0702895777522644e-05,
      "loss": 1.5187,
      "step": 14000
    },
    {
      "epoch": 16.65,
      "learning_rate": 7.513713483862737e-06,
      "loss": 1.5187,
      "step": 14500
    },
    {
      "epoch": 17.0,
      "eval_loss": 2.0979349613189697,
      "eval_runtime": 45.9022,
      "eval_samples_per_second": 63.897,
      "eval_steps_per_second": 7.995,
      "step": 14807
    },
    {
      "epoch": 17.22,
      "learning_rate": 4.324531190202833e-06,
      "loss": 1.4819,
      "step": 15000
    },
    {
      "epoch": 17.8,
      "learning_rate": 1.1353488965429266e-06,
      "loss": 1.4809,
      "step": 15500
    },
    {
      "epoch": 18.0,
      "eval_loss": 2.0337767601013184,
      "eval_runtime": 45.9061,
      "eval_samples_per_second": 63.891,
      "eval_steps_per_second": 7.995,
      "step": 15678
    },
    {
      "epoch": 18.0,
      "step": 15678,
      "total_flos": 2.6405831822391706e+17,
      "train_loss": 1.8265393278187887,
      "train_runtime": 59534.5244,
      "train_samples_per_second": 16.852,
      "train_steps_per_second": 0.263
    }
  ],
  "max_steps": 15678,
  "num_train_epochs": 18,
  "total_flos": 2.6405831822391706e+17,
  "trial_name": null,
  "trial_params": null
}