{
  "best_metric": 0.1787187159061432,
  "best_model_checkpoint": "output_pipe/1/origin/checkpoint-1600",
  "epoch": 4.0,
  "eval_steps": 200,
  "global_step": 3372,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11862396204033215,
      "grad_norm": 1.8725565671920776,
      "learning_rate": 2.954846478025286e-05,
      "loss": 0.4472,
      "step": 100
    },
    {
      "epoch": 0.2372479240806643,
      "grad_norm": 1.1395673751831055,
      "learning_rate": 2.8645394340758582e-05,
      "loss": 0.2467,
      "step": 200
    },
    {
      "epoch": 0.2372479240806643,
      "eval_accuracy": 0.9154929577464789,
      "eval_f1": 0.915466645488529,
      "eval_loss": 0.20940396189689636,
      "eval_matthews_correlation": 0.8319137927575645,
      "eval_precision": 0.9162967806708147,
      "eval_recall": 0.9156172895842885,
      "eval_runtime": 1.6946,
      "eval_samples_per_second": 3980.23,
      "eval_steps_per_second": 62.551,
      "step": 200
    },
    {
      "epoch": 0.35587188612099646,
      "grad_norm": 1.8331480026245117,
      "learning_rate": 2.77423239012643e-05,
      "loss": 0.2189,
      "step": 300
    },
    {
      "epoch": 0.4744958481613286,
      "grad_norm": 1.3709731101989746,
      "learning_rate": 2.6839253461770017e-05,
      "loss": 0.2134,
      "step": 400
    },
    {
      "epoch": 0.4744958481613286,
      "eval_accuracy": 0.9232023721275019,
      "eval_f1": 0.9231381305409507,
      "eval_loss": 0.18485188484191895,
      "eval_matthews_correlation": 0.8484666467701403,
      "eval_precision": 0.925079584759203,
      "eval_recall": 0.923388746773836,
      "eval_runtime": 1.6874,
      "eval_samples_per_second": 3997.323,
      "eval_steps_per_second": 62.819,
      "step": 400
    },
    {
      "epoch": 0.5931198102016607,
      "grad_norm": 2.092278480529785,
      "learning_rate": 2.5936183022275738e-05,
      "loss": 0.1867,
      "step": 500
    },
    {
      "epoch": 0.7117437722419929,
      "grad_norm": 1.712715983390808,
      "learning_rate": 2.503311258278146e-05,
      "loss": 0.1997,
      "step": 600
    },
    {
      "epoch": 0.7117437722419929,
      "eval_accuracy": 0.9113417346182358,
      "eval_f1": 0.9109541470737577,
      "eval_loss": 0.2311764806509018,
      "eval_matthews_correlation": 0.8313385287518686,
      "eval_precision": 0.919645119905826,
      "eval_recall": 0.9117310774056058,
      "eval_runtime": 1.687,
      "eval_samples_per_second": 3998.28,
      "eval_steps_per_second": 62.834,
      "step": 600
    },
    {
      "epoch": 0.830367734282325,
      "grad_norm": 1.4122563600540161,
      "learning_rate": 2.413004214328718e-05,
      "loss": 0.195,
      "step": 700
    },
    {
      "epoch": 0.9489916963226572,
      "grad_norm": 1.1271497011184692,
      "learning_rate": 2.3226971703792897e-05,
      "loss": 0.1924,
      "step": 800
    },
    {
      "epoch": 0.9489916963226572,
      "eval_accuracy": 0.9270570793180133,
      "eval_f1": 0.927051869770923,
      "eval_loss": 0.18776202201843262,
      "eval_matthews_correlation": 0.8541200435177229,
      "eval_precision": 0.9270814262829052,
      "eval_recall": 0.9270386183075724,
      "eval_runtime": 1.685,
      "eval_samples_per_second": 4002.966,
      "eval_steps_per_second": 62.908,
      "step": 800
    },
    {
      "epoch": 1.0676156583629894,
      "grad_norm": 1.3544421195983887,
      "learning_rate": 2.2323901264298614e-05,
      "loss": 0.1572,
      "step": 900
    },
    {
      "epoch": 1.1862396204033214,
      "grad_norm": 1.1566143035888672,
      "learning_rate": 2.1420830824804335e-05,
      "loss": 0.1152,
      "step": 1000
    },
    {
      "epoch": 1.1862396204033214,
      "eval_accuracy": 0.9298739807264641,
      "eval_f1": 0.9298428861607735,
      "eval_loss": 0.17949999868869781,
      "eval_matthews_correlation": 0.8609975455341051,
      "eval_precision": 0.9309806330562682,
      "eval_recall": 0.9300174512244082,
      "eval_runtime": 1.6857,
      "eval_samples_per_second": 4001.288,
      "eval_steps_per_second": 62.882,
      "step": 1000
    },
    {
      "epoch": 1.3048635824436536,
      "grad_norm": 2.370692014694214,
      "learning_rate": 2.0517760385310056e-05,
      "loss": 0.121,
      "step": 1100
    },
    {
      "epoch": 1.4234875444839858,
      "grad_norm": 1.542561411857605,
      "learning_rate": 1.9614689945815773e-05,
      "loss": 0.1101,
      "step": 1200
    },
    {
      "epoch": 1.4234875444839858,
      "eval_accuracy": 0.9277983691623425,
      "eval_f1": 0.9277947124771895,
      "eval_loss": 0.1969885528087616,
      "eval_matthews_correlation": 0.8558841839706139,
      "eval_precision": 0.9280188707623507,
      "eval_recall": 0.9278653269809751,
      "eval_runtime": 1.6846,
      "eval_samples_per_second": 4003.886,
      "eval_steps_per_second": 62.922,
      "step": 1200
    },
    {
      "epoch": 1.5421115065243178,
      "grad_norm": 4.452613353729248,
      "learning_rate": 1.8711619506321494e-05,
      "loss": 0.1136,
      "step": 1300
    },
    {
      "epoch": 1.66073546856465,
      "grad_norm": 2.4608500003814697,
      "learning_rate": 1.780854906682721e-05,
      "loss": 0.1125,
      "step": 1400
    },
    {
      "epoch": 1.66073546856465,
      "eval_accuracy": 0.9255744996293551,
      "eval_f1": 0.9255711867759875,
      "eval_loss": 0.20541010797023773,
      "eval_matthews_correlation": 0.8511449778675413,
      "eval_precision": 0.9255792243666476,
      "eval_recall": 0.9255657536074919,
      "eval_runtime": 1.6859,
      "eval_samples_per_second": 4000.789,
      "eval_steps_per_second": 62.874,
      "step": 1400
    },
    {
      "epoch": 1.7793594306049823,
      "grad_norm": 2.1233558654785156,
      "learning_rate": 1.6905478627332932e-05,
      "loss": 0.1088,
      "step": 1500
    },
    {
      "epoch": 1.8979833926453145,
      "grad_norm": 1.0472452640533447,
      "learning_rate": 1.6002408187838653e-05,
      "loss": 0.106,
      "step": 1600
    },
    {
      "epoch": 1.8979833926453145,
      "eval_accuracy": 0.9291326908821349,
      "eval_f1": 0.9290999252320007,
      "eval_loss": 0.1787187159061432,
      "eval_matthews_correlation": 0.8595549598842581,
      "eval_precision": 0.9302769607255612,
      "eval_recall": 0.9292785789737364,
      "eval_runtime": 1.6867,
      "eval_samples_per_second": 3999.043,
      "eval_steps_per_second": 62.846,
      "step": 1600
    },
    {
      "epoch": 2.0166073546856467,
      "grad_norm": 1.0865042209625244,
      "learning_rate": 1.5099337748344372e-05,
      "loss": 0.0951,
      "step": 1700
    },
    {
      "epoch": 2.135231316725979,
      "grad_norm": 0.06551714986562729,
      "learning_rate": 1.4196267308850091e-05,
      "loss": 0.0291,
      "step": 1800
    },
    {
      "epoch": 2.135231316725979,
      "eval_accuracy": 0.9310600444773907,
      "eval_f1": 0.9310172492672181,
      "eval_loss": 0.27395206689834595,
      "eval_matthews_correlation": 0.8637581637880172,
      "eval_precision": 0.9325344603488817,
      "eval_recall": 0.9312246964719653,
      "eval_runtime": 1.6876,
      "eval_samples_per_second": 3996.916,
      "eval_steps_per_second": 62.813,
      "step": 1800
    },
    {
      "epoch": 2.2538552787663106,
      "grad_norm": 0.18711210787296295,
      "learning_rate": 1.3302227573750754e-05,
      "loss": 0.0221,
      "step": 1900
    },
    {
      "epoch": 2.372479240806643,
      "grad_norm": 6.022641181945801,
      "learning_rate": 1.2399157134256473e-05,
      "loss": 0.0314,
      "step": 2000
    },
    {
      "epoch": 2.372479240806643,
      "eval_accuracy": 0.9269088213491475,
      "eval_f1": 0.9269088213491475,
      "eval_loss": 0.3283500075340271,
      "eval_matthews_correlation": 0.8538734279786175,
      "eval_precision": 0.9269367139893088,
      "eval_recall": 0.9269367139893088,
      "eval_runtime": 1.6876,
      "eval_samples_per_second": 3996.72,
      "eval_steps_per_second": 62.81,
      "step": 2000
    },
    {
      "epoch": 2.491103202846975,
      "grad_norm": 1.6997628211975098,
      "learning_rate": 1.1496086694762192e-05,
      "loss": 0.0269,
      "step": 2100
    },
    {
      "epoch": 2.6097271648873073,
      "grad_norm": 0.4570225179195404,
      "learning_rate": 1.0593016255267911e-05,
      "loss": 0.0314,
      "step": 2200
    },
    {
      "epoch": 2.6097271648873073,
      "eval_accuracy": 0.9313565604151223,
      "eval_f1": 0.9313562646866966,
      "eval_loss": 0.27509990334510803,
      "eval_matthews_correlation": 0.8628190296447534,
      "eval_precision": 0.9314230569697609,
      "eval_recall": 0.9313959731000736,
      "eval_runtime": 1.6863,
      "eval_samples_per_second": 3999.955,
      "eval_steps_per_second": 62.861,
      "step": 2200
    },
    {
      "epoch": 2.7283511269276395,
      "grad_norm": 0.349313348531723,
      "learning_rate": 9.68994581577363e-06,
      "loss": 0.0191,
      "step": 2300
    },
    {
      "epoch": 2.8469750889679717,
      "grad_norm": 0.8307677507400513,
      "learning_rate": 8.78687537627935e-06,
      "loss": 0.02,
      "step": 2400
    },
    {
      "epoch": 2.8469750889679717,
      "eval_accuracy": 0.9258710155670867,
      "eval_f1": 0.9258623315486654,
      "eval_loss": 0.31465885043144226,
      "eval_matthews_correlation": 0.8522173954984631,
      "eval_precision": 0.926259200744999,
      "eval_recall": 0.9259582478928491,
      "eval_runtime": 1.6874,
      "eval_samples_per_second": 3997.242,
      "eval_steps_per_second": 62.818,
      "step": 2400
    },
    {
      "epoch": 2.9655990510083035,
      "grad_norm": 4.751429080963135,
      "learning_rate": 7.883804936785068e-06,
      "loss": 0.0257,
      "step": 2500
    },
    {
      "epoch": 3.0842230130486357,
      "grad_norm": 0.01179005578160286,
      "learning_rate": 6.980734497290789e-06,
      "loss": 0.0098,
      "step": 2600
    },
    {
      "epoch": 3.0842230130486357,
      "eval_accuracy": 0.9279466271312083,
      "eval_f1": 0.9279466255474444,
      "eval_loss": 0.3863048255443573,
      "eval_matthews_correlation": 0.85595215852554,
      "eval_precision": 0.9279767942253909,
      "eval_recall": 0.9279753643013435,
      "eval_runtime": 1.6868,
      "eval_samples_per_second": 3998.746,
      "eval_steps_per_second": 62.842,
      "step": 2600
    },
    {
      "epoch": 3.202846975088968,
      "grad_norm": 0.02915983274579048,
      "learning_rate": 6.077664057796508e-06,
      "loss": 0.0046,
      "step": 2700
    },
    {
      "epoch": 3.3214709371293,
      "grad_norm": 0.011341964825987816,
      "learning_rate": 5.174593618302228e-06,
      "loss": 0.0043,
      "step": 2800
    },
    {
      "epoch": 3.3214709371293,
      "eval_accuracy": 0.9272053372868791,
      "eval_f1": 0.927201336922934,
      "eval_loss": 0.4001430571079254,
      "eval_matthews_correlation": 0.8547112744518932,
      "eval_precision": 0.9274373862263419,
      "eval_recall": 0.9272739038603537,
      "eval_runtime": 1.6874,
      "eval_samples_per_second": 3997.383,
      "eval_steps_per_second": 62.82,
      "step": 2800
    },
    {
      "epoch": 3.4400948991696323,
      "grad_norm": 0.10908253490924835,
      "learning_rate": 4.271523178807947e-06,
      "loss": 0.0057,
      "step": 2900
    },
    {
      "epoch": 3.5587188612099645,
      "grad_norm": 0.01483425498008728,
      "learning_rate": 3.3684527393136663e-06,
      "loss": 0.006,
      "step": 3000
    },
    {
      "epoch": 3.5587188612099645,
      "eval_accuracy": 0.9291326908821349,
      "eval_f1": 0.9291256977055623,
      "eval_loss": 0.37901368737220764,
      "eval_matthews_correlation": 0.8586933656020669,
      "eval_precision": 0.929478264358094,
      "eval_recall": 0.9292151415573195,
      "eval_runtime": 1.6857,
      "eval_samples_per_second": 4001.304,
      "eval_steps_per_second": 62.882,
      "step": 3000
    },
    {
      "epoch": 3.6773428232502967,
      "grad_norm": 0.04719650372862816,
      "learning_rate": 2.465382299819386e-06,
      "loss": 0.0019,
      "step": 3100
    },
    {
      "epoch": 3.795966785290629,
      "grad_norm": 0.04005097970366478,
      "learning_rate": 1.5623118603251053e-06,
      "loss": 0.0041,
      "step": 3200
    },
    {
      "epoch": 3.795966785290629,
      "eval_accuracy": 0.9285396590066716,
      "eval_f1": 0.9285317400859056,
      "eval_loss": 0.414096862077713,
      "eval_matthews_correlation": 0.8575398171568404,
      "eval_precision": 0.9289145208770282,
      "eval_recall": 0.928625345037119,
      "eval_runtime": 1.6873,
      "eval_samples_per_second": 3997.603,
      "eval_steps_per_second": 62.824,
      "step": 3200
    },
    {
      "epoch": 3.914590747330961,
      "grad_norm": 16.630456924438477,
      "learning_rate": 6.592414208308247e-07,
      "loss": 0.002,
      "step": 3300
    },
    {
      "epoch": 4.0,
      "step": 3372,
      "total_flos": 2.400334555212288e+16,
      "train_loss": 0.09447565742393829,
      "train_runtime": 348.9444,
      "train_samples_per_second": 618.46,
      "train_steps_per_second": 9.663
    }
  ],
  "logging_steps": 100,
  "max_steps": 3372,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.400334555212288e+16,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}