{
  "best_metric": 0.46305870034683594,
  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-11910",
  "epoch": 9.997061416397296,
  "eval_steps": 500,
  "global_step": 17010,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.29385836027034967,
      "grad_norm": 1.9668159484863281,
      "learning_rate": 4.853027630805409e-05,
      "loss": 0.4174,
      "step": 500
    },
    {
      "epoch": 0.5877167205406993,
      "grad_norm": 3.246731758117676,
      "learning_rate": 4.7060552616108174e-05,
      "loss": 0.2765,
      "step": 1000
    },
    {
      "epoch": 0.8815750808110491,
      "grad_norm": 2.936720609664917,
      "learning_rate": 4.559082892416226e-05,
      "loss": 0.2596,
      "step": 1500
    },
    {
      "epoch": 0.9997061416397296,
      "eval_accuracy": 0.8358953937837708,
      "eval_f1": 0.3926881971617123,
      "eval_loss": 0.431875616312027,
      "eval_precision": 0.26165460347108727,
      "eval_recall": 0.786616752456715,
      "eval_runtime": 16.2639,
      "eval_samples_per_second": 418.535,
      "eval_steps_per_second": 52.325,
      "step": 1701
    },
    {
      "epoch": 1.1754334410813987,
      "grad_norm": 2.302980661392212,
      "learning_rate": 4.4121105232216346e-05,
      "loss": 0.2135,
      "step": 2000
    },
    {
      "epoch": 1.4692918013517484,
      "grad_norm": 3.7150967121124268,
      "learning_rate": 4.265138154027043e-05,
      "loss": 0.1839,
      "step": 2500
    },
    {
      "epoch": 1.7631501616220981,
      "grad_norm": 1.2019191980361938,
      "learning_rate": 4.118165784832452e-05,
      "loss": 0.1853,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8644911466450625,
      "eval_f1": 0.44846535316981634,
      "eval_loss": 0.3841294050216675,
      "eval_precision": 0.3142374154770849,
      "eval_recall": 0.7828731867103416,
      "eval_runtime": 16.2484,
      "eval_samples_per_second": 418.933,
      "eval_steps_per_second": 52.374,
      "step": 3403
    },
    {
      "epoch": 2.0570085218924477,
      "grad_norm": 2.090308904647827,
      "learning_rate": 3.971193415637861e-05,
      "loss": 0.1684,
      "step": 3500
    },
    {
      "epoch": 2.3508668821627974,
      "grad_norm": 0.9995286464691162,
      "learning_rate": 3.824221046443269e-05,
      "loss": 0.1222,
      "step": 4000
    },
    {
      "epoch": 2.644725242433147,
      "grad_norm": 1.3526027202606201,
      "learning_rate": 3.677248677248677e-05,
      "loss": 0.1277,
      "step": 4500
    },
    {
      "epoch": 2.938583602703497,
      "grad_norm": 2.2816500663757324,
      "learning_rate": 3.530276308054086e-05,
      "loss": 0.1254,
      "step": 5000
    },
    {
      "epoch": 2.9997061416397295,
      "eval_accuracy": 0.843626202360437,
      "eval_f1": 0.4435178651613316,
      "eval_loss": 0.6409665942192078,
      "eval_precision": 0.30552364118426867,
      "eval_recall": 0.8088441740758072,
      "eval_runtime": 16.2306,
      "eval_samples_per_second": 419.394,
      "eval_steps_per_second": 52.432,
      "step": 5104
    },
    {
      "epoch": 3.2324419629738466,
      "grad_norm": 0.6440290808677673,
      "learning_rate": 3.3833039388594945e-05,
      "loss": 0.0929,
      "step": 5500
    },
    {
      "epoch": 3.5263003232441963,
      "grad_norm": 3.7906153202056885,
      "learning_rate": 3.2363315696649034e-05,
      "loss": 0.0896,
      "step": 6000
    },
    {
      "epoch": 3.820158683514546,
      "grad_norm": 1.0953032970428467,
      "learning_rate": 3.0893592004703116e-05,
      "loss": 0.0823,
      "step": 6500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.843598739452527,
      "eval_f1": 0.4336244267135767,
      "eval_loss": 0.7241775393486023,
      "eval_precision": 0.2964012711500472,
      "eval_recall": 0.8074403369209172,
      "eval_runtime": 16.5937,
      "eval_samples_per_second": 410.216,
      "eval_steps_per_second": 51.285,
      "step": 6806
    },
    {
      "epoch": 4.114017043784895,
      "grad_norm": 1.7823286056518555,
      "learning_rate": 2.9423868312757202e-05,
      "loss": 0.0761,
      "step": 7000
    },
    {
      "epoch": 4.407875404055245,
      "grad_norm": 1.165720820426941,
      "learning_rate": 2.795414462081129e-05,
      "loss": 0.0603,
      "step": 7500
    },
    {
      "epoch": 4.701733764325595,
      "grad_norm": 4.1130452156066895,
      "learning_rate": 2.648442092886537e-05,
      "loss": 0.0589,
      "step": 8000
    },
    {
      "epoch": 4.9955921245959445,
      "grad_norm": 0.3957385718822479,
      "learning_rate": 2.501469723691946e-05,
      "loss": 0.0597,
      "step": 8500
    },
    {
      "epoch": 4.99970614163973,
      "eval_accuracy": 0.8501829716239504,
      "eval_f1": 0.4494872643069798,
      "eval_loss": 0.7755796909332275,
      "eval_precision": 0.3133474771700028,
      "eval_recall": 0.7948058025269069,
      "eval_runtime": 16.1833,
      "eval_samples_per_second": 420.619,
      "eval_steps_per_second": 52.585,
      "step": 8507
    },
    {
      "epoch": 5.289450484866294,
      "grad_norm": 1.0087623596191406,
      "learning_rate": 2.3544973544973546e-05,
      "loss": 0.0423,
      "step": 9000
    },
    {
      "epoch": 5.583308845136644,
      "grad_norm": 2.162200450897217,
      "learning_rate": 2.2075249853027632e-05,
      "loss": 0.043,
      "step": 9500
    },
    {
      "epoch": 5.877167205406994,
      "grad_norm": 1.1820895671844482,
      "learning_rate": 2.0605526161081718e-05,
      "loss": 0.0446,
      "step": 10000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8483223596130476,
      "eval_f1": 0.4512908099586152,
      "eval_loss": 0.8560824394226074,
      "eval_precision": 0.31372728103023106,
      "eval_recall": 0.8036967711745437,
      "eval_runtime": 16.2581,
      "eval_samples_per_second": 418.684,
      "eval_steps_per_second": 52.343,
      "step": 10209
    },
    {
      "epoch": 6.171025565677343,
      "grad_norm": 2.190476894378662,
      "learning_rate": 1.91358024691358e-05,
      "loss": 0.0333,
      "step": 10500
    },
    {
      "epoch": 6.464883925947693,
      "grad_norm": 1.5180469751358032,
      "learning_rate": 1.766607877718989e-05,
      "loss": 0.0333,
      "step": 11000
    },
    {
      "epoch": 6.758742286218043,
      "grad_norm": 1.4279770851135254,
      "learning_rate": 1.6196355085243976e-05,
      "loss": 0.0325,
      "step": 11500
    },
    {
      "epoch": 6.99970614163973,
      "eval_accuracy": 0.8559776451929613,
      "eval_f1": 0.46305870034683594,
      "eval_loss": 0.9498798847198486,
      "eval_precision": 0.32143181611701643,
      "eval_recall": 0.8277959756668226,
      "eval_runtime": 16.1891,
      "eval_samples_per_second": 420.467,
      "eval_steps_per_second": 52.566,
      "step": 11910
    },
    {
      "epoch": 7.052600646488393,
      "grad_norm": 0.6558970212936401,
      "learning_rate": 1.472663139329806e-05,
      "loss": 0.0289,
      "step": 12000
    },
    {
      "epoch": 7.346459006758742,
      "grad_norm": 0.26189878582954407,
      "learning_rate": 1.3256907701352148e-05,
      "loss": 0.0224,
      "step": 12500
    },
    {
      "epoch": 7.640317367029092,
      "grad_norm": 1.370686650276184,
      "learning_rate": 1.1787184009406232e-05,
      "loss": 0.0231,
      "step": 13000
    },
    {
      "epoch": 7.934175727299442,
      "grad_norm": 0.36619672179222107,
      "learning_rate": 1.0317460317460318e-05,
      "loss": 0.022,
      "step": 13500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.851041187496138,
      "eval_f1": 0.45333161323614785,
      "eval_loss": 1.045249104499817,
      "eval_precision": 0.3129397096802921,
      "eval_recall": 0.8221806270472625,
      "eval_runtime": 16.479,
      "eval_samples_per_second": 413.071,
      "eval_steps_per_second": 51.641,
      "step": 13612
    },
    {
      "epoch": 8.22803408756979,
      "grad_norm": 1.0557399988174438,
      "learning_rate": 8.847736625514404e-06,
      "loss": 0.0164,
      "step": 14000
    },
    {
      "epoch": 8.521892447840141,
      "grad_norm": 4.32920503616333,
      "learning_rate": 7.37801293356849e-06,
      "loss": 0.0164,
      "step": 14500
    },
    {
      "epoch": 8.81575080811049,
      "grad_norm": 0.1674884408712387,
      "learning_rate": 5.908289241622575e-06,
      "loss": 0.017,
      "step": 15000
    },
    {
      "epoch": 8.999706141639729,
      "eval_accuracy": 0.8523937357107058,
      "eval_f1": 0.45305514157973176,
      "eval_loss": 1.102479100227356,
      "eval_precision": 0.31328972130119187,
      "eval_recall": 0.8179691155825924,
      "eval_runtime": 16.1914,
      "eval_samples_per_second": 420.409,
      "eval_steps_per_second": 52.559,
      "step": 15313
    },
    {
      "epoch": 9.10960916838084,
      "grad_norm": 3.3734261989593506,
      "learning_rate": 4.438565549676661e-06,
      "loss": 0.0171,
      "step": 15500
    },
    {
      "epoch": 9.40346752865119,
      "grad_norm": 0.19828377664089203,
      "learning_rate": 2.9688418577307467e-06,
      "loss": 0.013,
      "step": 16000
    },
    {
      "epoch": 9.69732588892154,
      "grad_norm": 1.491190791130066,
      "learning_rate": 1.4991181657848325e-06,
      "loss": 0.0133,
      "step": 16500
    },
    {
      "epoch": 9.991184249191889,
      "grad_norm": 2.5641109943389893,
      "learning_rate": 2.9394473838918286e-08,
      "loss": 0.0135,
      "step": 17000
    },
    {
      "epoch": 9.997061416397296,
      "eval_accuracy": 0.8526409018818958,
      "eval_f1": 0.45504563402161957,
      "eval_loss": 1.1188451051712036,
      "eval_precision": 0.3145413870246085,
      "eval_recall": 0.8224145999064109,
      "eval_runtime": 16.3165,
      "eval_samples_per_second": 417.185,
      "eval_steps_per_second": 52.156,
      "step": 17010
    },
    {
      "epoch": 9.997061416397296,
      "step": 17010,
      "total_flos": 6700722040732752.0,
      "train_loss": 0.08913132946046923,
      "train_runtime": 3311.208,
      "train_samples_per_second": 82.218,
      "train_steps_per_second": 5.137
    }
  ],
  "logging_steps": 500,
  "max_steps": 17010,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 6700722040732752.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}