{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.666666666666667,
  "eval_steps": 10,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "learning_rate": 4.9198396793587176e-05,
      "loss": 1.4613,
      "step": 10
    },
    {
      "epoch": 0.13,
      "eval_loss": 1.749655842781067,
      "eval_runtime": 9.1057,
      "eval_samples_per_second": 0.988,
      "eval_steps_per_second": 0.22,
      "step": 10
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.829659318637275e-05,
      "loss": 0.8073,
      "step": 20
    },
    {
      "epoch": 0.27,
      "eval_loss": 1.7181354761123657,
      "eval_runtime": 9.0405,
      "eval_samples_per_second": 0.996,
      "eval_steps_per_second": 0.221,
      "step": 20
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.7294589178356715e-05,
      "loss": 0.5993,
      "step": 30
    },
    {
      "epoch": 0.4,
      "eval_loss": 1.6926982402801514,
      "eval_runtime": 9.0635,
      "eval_samples_per_second": 0.993,
      "eval_steps_per_second": 0.221,
      "step": 30
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.629258517034069e-05,
      "loss": 0.5535,
      "step": 40
    },
    {
      "epoch": 0.53,
      "eval_loss": 1.5878939628601074,
      "eval_runtime": 9.0858,
      "eval_samples_per_second": 0.991,
      "eval_steps_per_second": 0.22,
      "step": 40
    },
    {
      "epoch": 0.67,
      "learning_rate": 4.529058116232465e-05,
      "loss": 0.4914,
      "step": 50
    },
    {
      "epoch": 0.67,
      "eval_loss": 1.6263775825500488,
      "eval_runtime": 9.0441,
      "eval_samples_per_second": 0.995,
      "eval_steps_per_second": 0.221,
      "step": 50
    },
    {
      "epoch": 0.8,
      "learning_rate": 4.428857715430862e-05,
      "loss": 0.3532,
      "step": 60
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.5376472473144531,
      "eval_runtime": 9.0097,
      "eval_samples_per_second": 0.999,
      "eval_steps_per_second": 0.222,
      "step": 60
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.3286573146292584e-05,
      "loss": 0.3079,
      "step": 70
    },
    {
      "epoch": 0.93,
      "eval_loss": 1.50021231174469,
      "eval_runtime": 9.0392,
      "eval_samples_per_second": 0.996,
      "eval_steps_per_second": 0.221,
      "step": 70
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.228456913827655e-05,
      "loss": 0.32,
      "step": 80
    },
    {
      "epoch": 1.07,
      "eval_loss": 1.4650906324386597,
      "eval_runtime": 9.0603,
      "eval_samples_per_second": 0.993,
      "eval_steps_per_second": 0.221,
      "step": 80
    },
    {
      "epoch": 1.2,
      "learning_rate": 4.128256513026052e-05,
      "loss": 0.22,
      "step": 90
    },
    {
      "epoch": 1.2,
      "eval_loss": 1.5176193714141846,
      "eval_runtime": 9.0417,
      "eval_samples_per_second": 0.995,
      "eval_steps_per_second": 0.221,
      "step": 90
    },
    {
      "epoch": 1.33,
      "learning_rate": 4.0280561122244495e-05,
      "loss": 0.2341,
      "step": 100
    },
    {
      "epoch": 1.33,
      "eval_loss": 1.4975649118423462,
      "eval_runtime": 9.0642,
      "eval_samples_per_second": 0.993,
      "eval_steps_per_second": 0.221,
      "step": 100
    },
    {
      "epoch": 1.47,
      "learning_rate": 3.927855711422846e-05,
      "loss": 0.2391,
      "step": 110
    },
    {
      "epoch": 1.47,
      "eval_loss": 1.7257053852081299,
      "eval_runtime": 8.9861,
      "eval_samples_per_second": 1.002,
      "eval_steps_per_second": 0.223,
      "step": 110
    },
    {
      "epoch": 1.6,
      "learning_rate": 3.8276553106212426e-05,
      "loss": 0.1884,
      "step": 120
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.639228105545044,
      "eval_runtime": 9.0813,
      "eval_samples_per_second": 0.991,
      "eval_steps_per_second": 0.22,
      "step": 120
    },
    {
      "epoch": 1.73,
      "learning_rate": 3.727454909819639e-05,
      "loss": 0.1881,
      "step": 130
    },
    {
      "epoch": 1.73,
      "eval_loss": 1.5887799263000488,
      "eval_runtime": 9.08,
      "eval_samples_per_second": 0.991,
      "eval_steps_per_second": 0.22,
      "step": 130
    },
    {
      "epoch": 1.87,
      "learning_rate": 3.627254509018036e-05,
      "loss": 0.1314,
      "step": 140
    },
    {
      "epoch": 1.87,
      "eval_loss": 1.6493072509765625,
      "eval_runtime": 9.0781,
      "eval_samples_per_second": 0.991,
      "eval_steps_per_second": 0.22,
      "step": 140
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.527054108216433e-05,
      "loss": 0.1524,
      "step": 150
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.6057649850845337,
      "eval_runtime": 9.0159,
      "eval_samples_per_second": 0.998,
      "eval_steps_per_second": 0.222,
      "step": 150
    },
    {
      "epoch": 2.13,
      "learning_rate": 3.42685370741483e-05,
      "loss": 0.1036,
      "step": 160
    },
    {
      "epoch": 2.13,
      "eval_loss": 1.6697440147399902,
      "eval_runtime": 8.9746,
      "eval_samples_per_second": 1.003,
      "eval_steps_per_second": 0.223,
      "step": 160
    },
    {
      "epoch": 2.27,
      "learning_rate": 3.326653306613227e-05,
      "loss": 0.1176,
      "step": 170
    },
    {
      "epoch": 2.27,
      "eval_loss": 1.6768882274627686,
      "eval_runtime": 9.0646,
      "eval_samples_per_second": 0.993,
      "eval_steps_per_second": 0.221,
      "step": 170
    },
    {
      "epoch": 2.4,
      "learning_rate": 3.2264529058116233e-05,
      "loss": 0.1375,
      "step": 180
    },
    {
      "epoch": 2.4,
      "eval_loss": 1.617688775062561,
      "eval_runtime": 9.0499,
      "eval_samples_per_second": 0.994,
      "eval_steps_per_second": 0.221,
      "step": 180
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.12625250501002e-05,
      "loss": 0.1038,
      "step": 190
    },
    {
      "epoch": 2.53,
      "eval_loss": 1.6981985569000244,
      "eval_runtime": 9.0561,
      "eval_samples_per_second": 0.994,
      "eval_steps_per_second": 0.221,
      "step": 190
    },
    {
      "epoch": 2.67,
      "learning_rate": 3.026052104208417e-05,
      "loss": 0.1068,
      "step": 200
    },
    {
      "epoch": 2.67,
      "eval_loss": 1.7827256917953491,
      "eval_runtime": 9.0294,
      "eval_samples_per_second": 0.997,
      "eval_steps_per_second": 0.221,
      "step": 200
    },
    {
      "epoch": 2.8,
      "learning_rate": 2.925851703406814e-05,
      "loss": 0.1229,
      "step": 210
    },
    {
      "epoch": 2.8,
      "eval_loss": 1.6968226432800293,
      "eval_runtime": 9.039,
      "eval_samples_per_second": 0.996,
      "eval_steps_per_second": 0.221,
      "step": 210
    },
    {
      "epoch": 2.93,
      "learning_rate": 2.8256513026052106e-05,
      "loss": 0.142,
      "step": 220
    },
    {
      "epoch": 2.93,
      "eval_loss": 1.7085565328598022,
      "eval_runtime": 9.0746,
      "eval_samples_per_second": 0.992,
      "eval_steps_per_second": 0.22,
      "step": 220
    },
    {
      "epoch": 3.07,
      "learning_rate": 2.7254509018036072e-05,
      "loss": 0.0868,
      "step": 230
    },
    {
      "epoch": 3.07,
      "eval_loss": 1.814492106437683,
      "eval_runtime": 9.0805,
      "eval_samples_per_second": 0.991,
      "eval_steps_per_second": 0.22,
      "step": 230
    },
    {
      "epoch": 3.2,
      "learning_rate": 2.625250501002004e-05,
      "loss": 0.0772,
      "step": 240
    },
    {
      "epoch": 3.2,
      "eval_loss": 1.916466236114502,
      "eval_runtime": 9.072,
      "eval_samples_per_second": 0.992,
      "eval_steps_per_second": 0.22,
      "step": 240
    },
    {
      "epoch": 3.33,
      "learning_rate": 2.5250501002004006e-05,
      "loss": 0.0644,
      "step": 250
    },
    {
      "epoch": 3.33,
      "eval_loss": 1.8792904615402222,
      "eval_runtime": 9.0616,
      "eval_samples_per_second": 0.993,
      "eval_steps_per_second": 0.221,
      "step": 250
    },
    {
      "epoch": 3.47,
      "learning_rate": 2.4248496993987975e-05,
      "loss": 0.0886,
      "step": 260
    },
    {
      "epoch": 3.47,
      "eval_loss": 1.7695975303649902,
      "eval_runtime": 9.0109,
      "eval_samples_per_second": 0.999,
      "eval_steps_per_second": 0.222,
      "step": 260
    },
    {
      "epoch": 3.6,
      "learning_rate": 2.3246492985971944e-05,
      "loss": 0.0807,
      "step": 270
    },
    {
      "epoch": 3.6,
      "eval_loss": 1.7431364059448242,
      "eval_runtime": 9.0009,
      "eval_samples_per_second": 1.0,
      "eval_steps_per_second": 0.222,
      "step": 270
    },
    {
      "epoch": 3.73,
      "learning_rate": 2.2244488977955913e-05,
      "loss": 0.0873,
      "step": 280
    },
    {
      "epoch": 3.73,
      "eval_loss": 1.8270233869552612,
      "eval_runtime": 9.0218,
      "eval_samples_per_second": 0.998,
      "eval_steps_per_second": 0.222,
      "step": 280
    },
    {
      "epoch": 3.87,
      "learning_rate": 2.124248496993988e-05,
      "loss": 0.0704,
      "step": 290
    },
    {
      "epoch": 3.87,
      "eval_loss": 1.8161801099777222,
      "eval_runtime": 9.0309,
      "eval_samples_per_second": 0.997,
      "eval_steps_per_second": 0.221,
      "step": 290
    },
    {
      "epoch": 4.0,
      "learning_rate": 2.0240480961923848e-05,
      "loss": 0.0729,
      "step": 300
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.8546494245529175,
      "eval_runtime": 9.0378,
      "eval_samples_per_second": 0.996,
      "eval_steps_per_second": 0.221,
      "step": 300
    },
    {
      "epoch": 4.13,
      "learning_rate": 1.9238476953907817e-05,
      "loss": 0.063,
      "step": 310
    },
    {
      "epoch": 4.13,
      "eval_loss": 1.9666298627853394,
      "eval_runtime": 9.008,
      "eval_samples_per_second": 0.999,
      "eval_steps_per_second": 0.222,
      "step": 310
    },
    {
      "epoch": 4.27,
      "learning_rate": 1.8236472945891783e-05,
      "loss": 0.0541,
      "step": 320
    },
    {
      "epoch": 4.27,
      "eval_loss": 1.9620505571365356,
      "eval_runtime": 9.0643,
      "eval_samples_per_second": 0.993,
      "eval_steps_per_second": 0.221,
      "step": 320
    },
    {
      "epoch": 4.4,
      "learning_rate": 1.7234468937875752e-05,
      "loss": 0.0648,
      "step": 330
    },
    {
      "epoch": 4.4,
      "eval_loss": 1.962950348854065,
      "eval_runtime": 9.0685,
      "eval_samples_per_second": 0.992,
      "eval_steps_per_second": 0.221,
      "step": 330
    },
    {
      "epoch": 4.53,
      "learning_rate": 1.623246492985972e-05,
      "loss": 0.0631,
      "step": 340
    },
    {
      "epoch": 4.53,
      "eval_loss": 1.9116477966308594,
      "eval_runtime": 9.0789,
      "eval_samples_per_second": 0.991,
      "eval_steps_per_second": 0.22,
      "step": 340
    },
    {
      "epoch": 4.67,
      "learning_rate": 1.523046092184369e-05,
      "loss": 0.0545,
      "step": 350
    },
    {
      "epoch": 4.67,
      "eval_loss": 1.9107557535171509,
      "eval_runtime": 9.0172,
      "eval_samples_per_second": 0.998,
      "eval_steps_per_second": 0.222,
      "step": 350
    },
    {
      "epoch": 4.8,
      "learning_rate": 1.4228456913827657e-05,
      "loss": 0.0622,
      "step": 360
    },
    {
      "epoch": 4.8,
      "eval_loss": 1.965379238128662,
      "eval_runtime": 8.9497,
      "eval_samples_per_second": 1.006,
      "eval_steps_per_second": 0.223,
      "step": 360
    },
    {
      "epoch": 4.93,
      "learning_rate": 1.3226452905811623e-05,
      "loss": 0.0523,
      "step": 370
    },
    {
      "epoch": 4.93,
      "eval_loss": 2.0281553268432617,
      "eval_runtime": 9.0358,
      "eval_samples_per_second": 0.996,
      "eval_steps_per_second": 0.221,
      "step": 370
    },
    {
      "epoch": 5.07,
      "learning_rate": 1.2224448897795592e-05,
      "loss": 0.0479,
      "step": 380
    },
    {
      "epoch": 5.07,
      "eval_loss": 2.065614938735962,
      "eval_runtime": 9.0421,
      "eval_samples_per_second": 0.995,
      "eval_steps_per_second": 0.221,
      "step": 380
    },
    {
      "epoch": 5.2,
      "learning_rate": 1.122244488977956e-05,
      "loss": 0.0373,
      "step": 390
    },
    {
      "epoch": 5.2,
      "eval_loss": 2.0752174854278564,
      "eval_runtime": 9.0688,
      "eval_samples_per_second": 0.992,
      "eval_steps_per_second": 0.221,
      "step": 390
    },
    {
      "epoch": 5.33,
      "learning_rate": 1.0220440881763528e-05,
      "loss": 0.0406,
      "step": 400
    },
    {
      "epoch": 5.33,
      "eval_loss": 2.0857064723968506,
      "eval_runtime": 9.0511,
      "eval_samples_per_second": 0.994,
      "eval_steps_per_second": 0.221,
      "step": 400
    },
    {
      "epoch": 5.47,
      "learning_rate": 9.218436873747496e-06,
      "loss": 0.0463,
      "step": 410
    },
    {
      "epoch": 5.47,
      "eval_loss": 2.118208885192871,
      "eval_runtime": 9.0064,
      "eval_samples_per_second": 0.999,
      "eval_steps_per_second": 0.222,
      "step": 410
    },
    {
      "epoch": 5.6,
      "learning_rate": 8.216432865731463e-06,
      "loss": 0.0433,
      "step": 420
    },
    {
      "epoch": 5.6,
      "eval_loss": 2.1232292652130127,
      "eval_runtime": 9.0773,
      "eval_samples_per_second": 0.991,
      "eval_steps_per_second": 0.22,
      "step": 420
    },
    {
      "epoch": 5.73,
      "learning_rate": 7.214428857715432e-06,
      "loss": 0.0425,
      "step": 430
    },
    {
      "epoch": 5.73,
      "eval_loss": 2.124908924102783,
      "eval_runtime": 9.0585,
      "eval_samples_per_second": 0.994,
      "eval_steps_per_second": 0.221,
      "step": 430
    },
    {
      "epoch": 5.87,
      "learning_rate": 6.212424849699399e-06,
      "loss": 0.0413,
      "step": 440
    },
    {
      "epoch": 5.87,
      "eval_loss": 2.125481605529785,
      "eval_runtime": 9.0338,
      "eval_samples_per_second": 0.996,
      "eval_steps_per_second": 0.221,
      "step": 440
    },
    {
      "epoch": 6.0,
      "learning_rate": 5.2104208416833665e-06,
      "loss": 0.0579,
      "step": 450
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.1075000762939453,
      "eval_runtime": 9.0414,
      "eval_samples_per_second": 0.995,
      "eval_steps_per_second": 0.221,
      "step": 450
    },
    {
      "epoch": 6.13,
      "learning_rate": 4.208416833667335e-06,
      "loss": 0.0326,
      "step": 460
    },
    {
      "epoch": 6.13,
      "eval_loss": 2.112290382385254,
      "eval_runtime": 8.9941,
      "eval_samples_per_second": 1.001,
      "eval_steps_per_second": 0.222,
      "step": 460
    },
    {
      "epoch": 6.27,
      "learning_rate": 3.2064128256513024e-06,
      "loss": 0.0378,
      "step": 470
    },
    {
      "epoch": 6.27,
      "eval_loss": 2.1354782581329346,
      "eval_runtime": 9.0542,
      "eval_samples_per_second": 0.994,
      "eval_steps_per_second": 0.221,
      "step": 470
    },
    {
      "epoch": 6.4,
      "learning_rate": 2.2044088176352706e-06,
      "loss": 0.0384,
      "step": 480
    },
    {
      "epoch": 6.4,
      "eval_loss": 2.1528055667877197,
      "eval_runtime": 9.0296,
      "eval_samples_per_second": 0.997,
      "eval_steps_per_second": 0.221,
      "step": 480
    },
    {
      "epoch": 6.53,
      "learning_rate": 1.2024048096192386e-06,
      "loss": 0.0328,
      "step": 490
    },
    {
      "epoch": 6.53,
      "eval_loss": 2.162899971008301,
      "eval_runtime": 9.0663,
      "eval_samples_per_second": 0.993,
      "eval_steps_per_second": 0.221,
      "step": 490
    },
    {
      "epoch": 6.67,
      "learning_rate": 2.004008016032064e-07,
      "loss": 0.0356,
      "step": 500
    },
    {
      "epoch": 6.67,
      "eval_loss": 2.1663544178009033,
      "eval_runtime": 9.0766,
      "eval_samples_per_second": 0.992,
      "eval_steps_per_second": 0.22,
      "step": 500
    }
  ],
  "logging_steps": 10,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 50,
  "total_flos": 3.411755398324224e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}