{
  "best_metric": 0.17586912065439672,
  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-3403",
  "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": 0.7671626010120082,
      "eval_f1": 0.14622946114658822,
      "eval_loss": 0.7913026213645935,
      "eval_precision": 0.0793057825511713,
      "eval_recall": 0.9365808823529411,
      "eval_runtime": 16.052,
      "eval_samples_per_second": 424.06,
      "eval_steps_per_second": 53.015,
      "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": 0.8099635429897495,
      "eval_f1": 0.17586912065439672,
      "eval_loss": 0.6631014347076416,
      "eval_precision": 0.09691960931630353,
      "eval_recall": 0.9485294117647058,
      "eval_runtime": 15.875,
      "eval_samples_per_second": 428.788,
      "eval_steps_per_second": 53.606,
      "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": 0.7754975935626944,
      "eval_f1": 0.16528259292106748,
      "eval_loss": 1.072888731956482,
      "eval_precision": 0.0905877154220062,
      "eval_recall": 0.9420955882352942,
      "eval_runtime": 15.8836,
      "eval_samples_per_second": 428.556,
      "eval_steps_per_second": 53.577,
      "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": 0.7719411469883488,
      "eval_f1": 0.16243814311523055,
      "eval_loss": 1.2567578554153442,
      "eval_precision": 0.08880872627329726,
      "eval_recall": 0.9503676470588235,
      "eval_runtime": 15.8976,
      "eval_samples_per_second": 428.179,
      "eval_steps_per_second": 53.53,
      "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": 0.7836540772119656,
      "eval_f1": 0.17101181993461315,
      "eval_loss": 1.1907650232315063,
      "eval_precision": 0.0940872613227562,
      "eval_recall": 0.9375,
      "eval_runtime": 16.019,
      "eval_samples_per_second": 424.932,
      "eval_steps_per_second": 53.124,
      "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": 0.7811686840461102,
      "eval_f1": 0.1718036055495555,
      "eval_loss": 1.384422779083252,
      "eval_precision": 0.09443784820531555,
      "eval_recall": 0.9503676470588235,
      "eval_runtime": 15.9006,
      "eval_samples_per_second": 428.098,
      "eval_steps_per_second": 53.52,
      "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": 0.7866406684471785,
      "eval_f1": 0.1704836709384043,
      "eval_loss": 1.5515447854995728,
      "eval_precision": 0.09366766603070772,
      "eval_recall": 0.9476102941176471,
      "eval_runtime": 15.8267,
      "eval_samples_per_second": 430.095,
      "eval_steps_per_second": 53.77,
      "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": 0.7842582611859856,
      "eval_f1": 0.1688722903304376,
      "eval_loss": 1.6299601793289185,
      "eval_precision": 0.09261733012734882,
      "eval_recall": 0.9558823529411765,
      "eval_runtime": 15.9057,
      "eval_samples_per_second": 427.961,
      "eval_steps_per_second": 53.503,
      "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": 0.7844848301762433,
      "eval_f1": 0.16934759532946844,
      "eval_loss": 1.7459304332733154,
      "eval_precision": 0.09292947396720136,
      "eval_recall": 0.953125,
      "eval_runtime": 16.0915,
      "eval_samples_per_second": 423.018,
      "eval_steps_per_second": 52.885,
      "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": 0.7845534874460183,
      "eval_f1": 0.16896354888689555,
      "eval_loss": 1.7860842943191528,
      "eval_precision": 0.09270693512304251,
      "eval_recall": 0.9522058823529411,
      "eval_runtime": 15.9319,
      "eval_samples_per_second": 427.256,
      "eval_steps_per_second": 53.415,
      "step": 17010
    },
    {
      "epoch": 9.997061416397296,
      "step": 17010,
      "total_flos": 6700722040732752.0,
      "train_loss": 0.08913132946046923,
      "train_runtime": 3337.712,
      "train_samples_per_second": 81.565,
      "train_steps_per_second": 5.096
    }
  ],
  "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
}