{ "best_metric": null, "best_model_checkpoint": null, "epoch": 100.0, "eval_steps": 500, "global_step": 9600, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 1.2695553302764893, "learning_rate": 4.9500000000000004e-05, "loss": 0.8815, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.0, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.0, "eval_LOCATION_recall": 0.0, "eval_ORGANIZATION_f1": 0.0, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.0, "eval_ORGANIZATION_recall": 0.0, "eval_PERSON_f1": 0.0, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.0, "eval_PERSON_recall": 0.0, "eval_loss": 0.5185229778289795, "eval_overall_accuracy": 0.8378900856117095, "eval_overall_f1": 0.0, "eval_overall_precision": 0.0, "eval_overall_recall": 0.0, "eval_runtime": 0.3004, "eval_samples_per_second": 565.841, "eval_steps_per_second": 9.985, "step": 96 }, { "epoch": 2.0, "grad_norm": 1.3206499814987183, "learning_rate": 4.9e-05, "loss": 0.3991, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.3644067796610169, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.3233082706766917, "eval_LOCATION_recall": 0.4174757281553398, "eval_ORGANIZATION_f1": 0.5203252032520325, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.48484848484848486, "eval_ORGANIZATION_recall": 0.5614035087719298, "eval_PERSON_f1": 0.5852090032154341, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.5055555555555555, "eval_PERSON_recall": 0.6946564885496184, "eval_loss": 0.2479129284620285, "eval_overall_accuracy": 0.923777961888981, "eval_overall_f1": 0.5021834061135371, "eval_overall_precision": 0.4500978473581213, "eval_overall_recall": 0.5679012345679012, "eval_runtime": 0.3016, "eval_samples_per_second": 563.742, "eval_steps_per_second": 9.948, "step": 192 }, { "epoch": 3.0, "grad_norm": 0.8349621891975403, "learning_rate": 4.85e-05, "loss": 0.2071, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.7106598984771573, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.7446808510638298, "eval_LOCATION_recall": 0.6796116504854369, "eval_ORGANIZATION_f1": 0.7966573816155988, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.7606382978723404, "eval_ORGANIZATION_recall": 0.8362573099415205, "eval_PERSON_f1": 0.962406015037594, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9481481481481482, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.1065019741654396, "eval_overall_accuracy": 0.9685169842584921, "eval_overall_f1": 0.829683698296837, "eval_overall_precision": 0.8177458033573142, "eval_overall_recall": 0.8419753086419753, "eval_runtime": 0.3021, "eval_samples_per_second": 562.671, "eval_steps_per_second": 9.929, "step": 288 }, { "epoch": 4.0, "grad_norm": 1.476486325263977, "learning_rate": 4.8e-05, "loss": 0.1278, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.848780487804878, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8529411764705882, "eval_LOCATION_recall": 0.8446601941747572, "eval_ORGANIZATION_f1": 0.8421052631578948, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9696969696969696, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9624060150375939, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.07688198983669281, "eval_overall_accuracy": 0.9762496547914941, "eval_overall_f1": 0.8843373493975903, "eval_overall_precision": 0.8635294117647059, "eval_overall_recall": 0.9061728395061729, "eval_runtime": 0.3017, "eval_samples_per_second": 563.433, "eval_steps_per_second": 9.943, "step": 384 }, { "epoch": 5.0, "grad_norm": 0.8666167855262756, "learning_rate": 4.75e-05, "loss": 0.1026, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.8725490196078431, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.8640776699029126, "eval_ORGANIZATION_f1": 0.8529411764705882, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8579881656804734, "eval_ORGANIZATION_recall": 0.847953216374269, "eval_PERSON_f1": 0.9548872180451129, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9407407407407408, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.06833094358444214, "eval_overall_accuracy": 0.9784589892294946, "eval_overall_f1": 0.891358024691358, "eval_overall_precision": 0.891358024691358, "eval_overall_recall": 0.891358024691358, "eval_runtime": 0.3066, "eval_samples_per_second": 554.454, "eval_steps_per_second": 9.784, "step": 480 }, { "epoch": 6.0, "grad_norm": 0.7837597131729126, "learning_rate": 4.7e-05, "loss": 0.087, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.8416289592760181, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.788135593220339, "eval_LOCATION_recall": 0.9029126213592233, "eval_ORGANIZATION_f1": 0.8606811145510835, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9144736842105263, "eval_ORGANIZATION_recall": 0.8128654970760234, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05511486530303955, "eval_overall_accuracy": 0.9814968240817453, "eval_overall_f1": 0.8921933085501859, "eval_overall_precision": 0.8955223880597015, "eval_overall_recall": 0.8888888888888888, "eval_runtime": 0.305, "eval_samples_per_second": 557.394, "eval_steps_per_second": 9.836, "step": 576 }, { "epoch": 7.0, "grad_norm": 1.0100133419036865, "learning_rate": 4.6500000000000005e-05, "loss": 0.0744, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.8855721393034826, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.8640776699029126, "eval_ORGANIZATION_f1": 0.9096209912536442, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9069767441860465, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.0486799031496048, "eval_overall_accuracy": 0.9853631593482464, "eval_overall_f1": 0.9244114002478315, "eval_overall_precision": 0.927860696517413, "eval_overall_recall": 0.9209876543209876, "eval_runtime": 0.3012, "eval_samples_per_second": 564.374, "eval_steps_per_second": 9.96, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.5667006969451904, "learning_rate": 4.600000000000001e-05, "loss": 0.0709, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.8727272727272728, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8205128205128205, "eval_LOCATION_recall": 0.9320388349514563, "eval_ORGANIZATION_f1": 0.8814589665653495, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9177215189873418, "eval_ORGANIZATION_recall": 0.847953216374269, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04678232595324516, "eval_overall_accuracy": 0.9845346589339962, "eval_overall_f1": 0.9064039408866995, "eval_overall_precision": 0.9041769041769042, "eval_overall_recall": 0.908641975308642, "eval_runtime": 0.3037, "eval_samples_per_second": 559.85, "eval_steps_per_second": 9.88, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.43462151288986206, "learning_rate": 4.55e-05, "loss": 0.0659, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.8909090909090909, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8376068376068376, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8727272727272727, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9056603773584906, "eval_ORGANIZATION_recall": 0.8421052631578947, "eval_PERSON_f1": 0.9660377358490565, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9552238805970149, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.045322172343730927, "eval_overall_accuracy": 0.9845346589339962, "eval_overall_f1": 0.9079754601226994, "eval_overall_precision": 0.9024390243902439, "eval_overall_recall": 0.9135802469135802, "eval_runtime": 0.3034, "eval_samples_per_second": 560.406, "eval_steps_per_second": 9.89, "step": 864 }, { "epoch": 10.0, "grad_norm": 0.7502378821372986, "learning_rate": 4.5e-05, "loss": 0.0584, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.9047619047619048, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8878504672897196, "eval_LOCATION_recall": 0.9223300970873787, "eval_ORGANIZATION_f1": 0.9154518950437317, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9127906976744186, "eval_ORGANIZATION_recall": 0.9181286549707602, "eval_PERSON_f1": 0.9696969696969696, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9624060150375939, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.040245313197374344, "eval_overall_accuracy": 0.9886771610052472, "eval_overall_f1": 0.9302325581395349, "eval_overall_precision": 0.9223300970873787, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.3011, "eval_samples_per_second": 564.581, "eval_steps_per_second": 9.963, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.5578929781913757, "learning_rate": 4.4500000000000004e-05, "loss": 0.0525, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.941747572815534, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.941747572815534, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9149560117302051, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9176470588235294, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.03952949866652489, "eval_overall_accuracy": 0.9884009942004971, "eval_overall_f1": 0.9407407407407408, "eval_overall_precision": 0.9407407407407408, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.301, "eval_samples_per_second": 564.723, "eval_steps_per_second": 9.966, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.5492796301841736, "learning_rate": 4.4000000000000006e-05, "loss": 0.0482, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.9201877934272301, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8909090909090909, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8955223880597015, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9146341463414634, "eval_ORGANIZATION_recall": 0.8771929824561403, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04383876919746399, "eval_overall_accuracy": 0.9861916597624966, "eval_overall_f1": 0.9236453201970444, "eval_overall_precision": 0.9213759213759214, "eval_overall_recall": 0.9259259259259259, "eval_runtime": 0.301, "eval_samples_per_second": 564.77, "eval_steps_per_second": 9.967, "step": 1152 }, { "epoch": 13.0, "grad_norm": 1.332250714302063, "learning_rate": 4.35e-05, "loss": 0.0441, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.9238095238095239, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9065420560747663, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.8967551622418879, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9047619047619048, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9696969696969696, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9624060150375939, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04452400282025337, "eval_overall_accuracy": 0.9861916597624966, "eval_overall_f1": 0.927429274292743, "eval_overall_precision": 0.9240196078431373, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.2996, "eval_samples_per_second": 567.33, "eval_steps_per_second": 10.012, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.2739567458629608, "learning_rate": 4.3e-05, "loss": 0.0401, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.909952606635071, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.9320388349514563, "eval_ORGANIZATION_f1": 0.8848484848484848, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9182389937106918, "eval_ORGANIZATION_recall": 0.8538011695906432, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04797976836562157, "eval_overall_accuracy": 0.9845346589339962, "eval_overall_f1": 0.9215442092154421, "eval_overall_precision": 0.9296482412060302, "eval_overall_recall": 0.9135802469135802, "eval_runtime": 0.3021, "eval_samples_per_second": 562.648, "eval_steps_per_second": 9.929, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.8639817237854004, "learning_rate": 4.25e-05, "loss": 0.0382, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.9333333333333335, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9158878504672897, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9090909090909091, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9117647058823529, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04100968316197395, "eval_overall_accuracy": 0.9884009942004971, "eval_overall_f1": 0.9372693726937269, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.3023, "eval_samples_per_second": 562.3, "eval_steps_per_second": 9.923, "step": 1440 }, { "epoch": 16.0, "grad_norm": 0.8146072626113892, "learning_rate": 4.2e-05, "loss": 0.0371, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.9245283018867926, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8990825688073395, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8967551622418879, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9047619047619048, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.042570844292640686, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.9262899262899262, "eval_overall_precision": 0.921760391198044, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.2996, "eval_samples_per_second": 567.508, "eval_steps_per_second": 10.015, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.34498077630996704, "learning_rate": 4.15e-05, "loss": 0.0323, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.9238095238095239, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9065420560747663, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9122807017543859, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9122807017543859, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04395383968949318, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9313725490196078, "eval_overall_precision": 0.9245742092457421, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.3015, "eval_samples_per_second": 563.808, "eval_steps_per_second": 9.95, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.4858402609825134, "learning_rate": 4.1e-05, "loss": 0.0304, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.9209302325581395, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8839285714285714, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9020771513353115, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04470779374241829, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9300613496932515, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.3003, "eval_samples_per_second": 566.176, "eval_steps_per_second": 9.991, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.33680614829063416, "learning_rate": 4.05e-05, "loss": 0.0277, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.8995215311004785, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8867924528301887, "eval_LOCATION_recall": 0.912621359223301, "eval_ORGANIZATION_f1": 0.8869047619047619, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9030303030303031, "eval_ORGANIZATION_recall": 0.8713450292397661, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04388991370797157, "eval_overall_accuracy": 0.9856393261529964, "eval_overall_f1": 0.9158415841584158, "eval_overall_precision": 0.9181141439205955, "eval_overall_recall": 0.9135802469135802, "eval_runtime": 0.3029, "eval_samples_per_second": 561.31, "eval_steps_per_second": 9.905, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.4714030921459198, "learning_rate": 4e-05, "loss": 0.0276, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.9238095238095239, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9065420560747663, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9032258064516129, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04655210301280022, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9312039312039312, "eval_overall_precision": 0.9266503667481663, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.3025, "eval_samples_per_second": 561.894, "eval_steps_per_second": 9.916, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.44346532225608826, "learning_rate": 3.9500000000000005e-05, "loss": 0.0264, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8761061946902655, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9057750759878419, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9430379746835443, "eval_ORGANIZATION_recall": 0.8713450292397661, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05079750716686249, "eval_overall_accuracy": 0.9853631593482464, "eval_overall_f1": 0.9282178217821783, "eval_overall_precision": 0.9305210918114144, "eval_overall_recall": 0.9259259259259259, "eval_runtime": 0.3007, "eval_samples_per_second": 565.368, "eval_steps_per_second": 9.977, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.45872893929481506, "learning_rate": 3.9000000000000006e-05, "loss": 0.0249, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.9345794392523364, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9166666666666666, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9333333333333333, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04766177013516426, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.939729397293973, "eval_overall_precision": 0.9362745098039216, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.2974, "eval_samples_per_second": 571.618, "eval_steps_per_second": 10.087, "step": 2112 }, { "epoch": 23.0, "grad_norm": 1.4570780992507935, "learning_rate": 3.85e-05, "loss": 0.0233, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.930232558139535, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8928571428571429, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9020771513353115, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9696969696969696, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9624060150375939, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04851089417934418, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9313725490196078, "eval_overall_precision": 0.9245742092457421, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2988, "eval_samples_per_second": 568.926, "eval_steps_per_second": 10.04, "step": 2208 }, { "epoch": 24.0, "grad_norm": 1.3498001098632812, "learning_rate": 3.8e-05, "loss": 0.0224, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9252336448598131, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8918918918918919, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9221556886227544, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9447852760736196, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9696969696969696, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9624060150375939, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04273885115981102, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.938423645320197, "eval_overall_precision": 0.9361179361179361, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.3006, "eval_samples_per_second": 565.475, "eval_steps_per_second": 9.979, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.38454458117485046, "learning_rate": 3.7500000000000003e-05, "loss": 0.0205, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.914027149321267, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8559322033898306, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.8996960486322189, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9367088607594937, "eval_ORGANIZATION_recall": 0.8654970760233918, "eval_PERSON_f1": 0.9660377358490565, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9552238805970149, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04676693305373192, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9251533742331289, "eval_overall_precision": 0.9195121951219513, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.2988, "eval_samples_per_second": 568.85, "eval_steps_per_second": 10.039, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.2634662389755249, "learning_rate": 3.7e-05, "loss": 0.0196, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9226190476190477, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9393939393939394, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04187300428748131, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9432098765432099, "eval_overall_precision": 0.9432098765432099, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.302, "eval_samples_per_second": 562.911, "eval_steps_per_second": 9.934, "step": 2496 }, { "epoch": 27.0, "grad_norm": 1.0896821022033691, "learning_rate": 3.65e-05, "loss": 0.0185, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9351851851851852, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8938053097345132, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.8914956011730205, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8941176470588236, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9696969696969696, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9624060150375939, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04213610664010048, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.928136419001218, "eval_overall_precision": 0.9158653846153846, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2984, "eval_samples_per_second": 569.76, "eval_steps_per_second": 10.055, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.6250720024108887, "learning_rate": 3.6e-05, "loss": 0.0183, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.951923076923077, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9428571428571428, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9164265129682997, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9034090909090909, "eval_ORGANIZATION_recall": 0.9298245614035088, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04632599651813507, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9437652811735942, "eval_overall_precision": 0.9346246973365617, "eval_overall_recall": 0.9530864197530864, "eval_runtime": 0.3, "eval_samples_per_second": 566.667, "eval_steps_per_second": 10.0, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.4535207450389862, "learning_rate": 3.55e-05, "loss": 0.0168, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9047619047619048, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8878504672897196, "eval_LOCATION_recall": 0.9223300970873787, "eval_ORGANIZATION_f1": 0.9074626865671643, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.926829268292683, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9545454545454546, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9473684210526315, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.04913368076086044, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9221260815822003, "eval_overall_precision": 0.9232673267326733, "eval_overall_recall": 0.9209876543209876, "eval_runtime": 0.3039, "eval_samples_per_second": 559.424, "eval_steps_per_second": 9.872, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.04675703123211861, "learning_rate": 3.5e-05, "loss": 0.0176, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.9230769230769231, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9142857142857143, "eval_LOCATION_recall": 0.9320388349514563, "eval_ORGANIZATION_f1": 0.8908045977011494, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8757062146892656, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04497746750712395, "eval_overall_accuracy": 0.9856393261529964, "eval_overall_f1": 0.9255189255189255, "eval_overall_precision": 0.9154589371980676, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2994, "eval_samples_per_second": 567.744, "eval_steps_per_second": 10.019, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.11599758267402649, "learning_rate": 3.45e-05, "loss": 0.0149, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.9201877934272301, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8909090909090909, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8882175226586103, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.91875, "eval_ORGANIZATION_recall": 0.8596491228070176, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05071546509861946, "eval_overall_accuracy": 0.9859154929577465, "eval_overall_f1": 0.9207920792079208, "eval_overall_precision": 0.9230769230769231, "eval_overall_recall": 0.9185185185185185, "eval_runtime": 0.2963, "eval_samples_per_second": 573.665, "eval_steps_per_second": 10.123, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.46507471799850464, "learning_rate": 3.4000000000000007e-05, "loss": 0.017, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9295774647887324, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9080118694362017, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9216867469879518, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04490618035197258, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9359605911330049, "eval_overall_precision": 0.9336609336609336, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2974, "eval_samples_per_second": 571.585, "eval_steps_per_second": 10.087, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.971469521522522, "learning_rate": 3.35e-05, "loss": 0.0142, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.9252336448598131, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8918918918918919, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.8842729970326408, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8975903614457831, "eval_ORGANIZATION_recall": 0.8713450292397661, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.045383553951978683, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9213759213759213, "eval_overall_precision": 0.9168704156479217, "eval_overall_recall": 0.9259259259259259, "eval_runtime": 0.299, "eval_samples_per_second": 568.603, "eval_steps_per_second": 10.034, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.46268317103385925, "learning_rate": 3.3e-05, "loss": 0.0142, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.9289099526066351, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9074074074074074, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8922155688622756, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9141104294478528, "eval_ORGANIZATION_recall": 0.8713450292397661, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05121282860636711, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9257425742574257, "eval_overall_precision": 0.9280397022332506, "eval_overall_recall": 0.9234567901234568, "eval_runtime": 0.2982, "eval_samples_per_second": 570.174, "eval_steps_per_second": 10.062, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.3609045743942261, "learning_rate": 3.2500000000000004e-05, "loss": 0.013, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.9439252336448598, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9099099099099099, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9024390243902439, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9426751592356688, "eval_ORGANIZATION_recall": 0.8654970760233918, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.0532350093126297, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9341614906832298, "eval_overall_precision": 0.94, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2993, "eval_samples_per_second": 567.954, "eval_steps_per_second": 10.023, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.35751500725746155, "learning_rate": 3.2000000000000005e-05, "loss": 0.0136, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.9252336448598131, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8918918918918919, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.8942598187311178, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.925, "eval_ORGANIZATION_recall": 0.8654970760233918, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05132954940199852, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9257425742574257, "eval_overall_precision": 0.9280397022332506, "eval_overall_recall": 0.9234567901234568, "eval_runtime": 0.2979, "eval_samples_per_second": 570.597, "eval_steps_per_second": 10.069, "step": 3456 }, { "epoch": 37.0, "grad_norm": 0.3856443464756012, "learning_rate": 3.15e-05, "loss": 0.012, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.9458128078817734, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.96, "eval_LOCATION_recall": 0.9320388349514563, "eval_ORGANIZATION_f1": 0.9117647058823529, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9171597633136095, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05015752464532852, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9379652605459058, "eval_overall_precision": 0.942643391521197, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2994, "eval_samples_per_second": 567.779, "eval_steps_per_second": 10.02, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.7762855291366577, "learning_rate": 3.1e-05, "loss": 0.0116, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9345794392523364, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.8942598187311178, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.925, "eval_ORGANIZATION_recall": 0.8654970760233918, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05322541669011116, "eval_overall_accuracy": 0.9856393261529964, "eval_overall_f1": 0.9282178217821783, "eval_overall_precision": 0.9305210918114144, "eval_overall_recall": 0.9259259259259259, "eval_runtime": 0.2995, "eval_samples_per_second": 567.66, "eval_steps_per_second": 10.018, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.3031044602394104, "learning_rate": 3.05e-05, "loss": 0.0113, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.9295774647887324, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.8928571428571428, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9090909090909091, "eval_ORGANIZATION_recall": 0.8771929824561403, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04845419526100159, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9297163995067819, "eval_overall_precision": 0.9285714285714286, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.2988, "eval_samples_per_second": 568.999, "eval_steps_per_second": 10.041, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.3121825158596039, "learning_rate": 3e-05, "loss": 0.0111, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.9289099526066351, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9074074074074074, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9053254437869822, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.051382340490818024, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9321824907521578, "eval_overall_precision": 0.9310344827586207, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2994, "eval_samples_per_second": 567.714, "eval_steps_per_second": 10.018, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.04050559177994728, "learning_rate": 2.95e-05, "loss": 0.0103, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.904191616766467, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05454375967383385, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9356435643564357, "eval_overall_precision": 0.9379652605459057, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2993, "eval_samples_per_second": 567.974, "eval_steps_per_second": 10.023, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.02801467664539814, "learning_rate": 2.9e-05, "loss": 0.0098, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.9428571428571428, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9252336448598131, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9032258064516129, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.052473269402980804, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9372693726937269, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2993, "eval_samples_per_second": 568.008, "eval_steps_per_second": 10.024, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.02515578828752041, "learning_rate": 2.8499999999999998e-05, "loss": 0.0102, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9282296650717704, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9150943396226415, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9154518950437317, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9127906976744186, "eval_ORGANIZATION_recall": 0.9181286549707602, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04892284795641899, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9349693251533743, "eval_overall_precision": 0.9292682926829269, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2988, "eval_samples_per_second": 568.921, "eval_steps_per_second": 10.04, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.13938526809215546, "learning_rate": 2.8000000000000003e-05, "loss": 0.0086, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9116279069767441, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.875, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8988095238095237, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06110783666372299, "eval_overall_accuracy": 0.9861916597624966, "eval_overall_f1": 0.927429274292743, "eval_overall_precision": 0.9240196078431373, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.2989, "eval_samples_per_second": 568.813, "eval_steps_per_second": 10.038, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.07607907056808472, "learning_rate": 2.7500000000000004e-05, "loss": 0.0099, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.9065420560747663, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8738738738738738, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9080118694362017, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9216867469879518, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.052707187831401825, "eval_overall_accuracy": 0.9859154929577465, "eval_overall_f1": 0.9262899262899262, "eval_overall_precision": 0.921760391198044, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.3057, "eval_samples_per_second": 556.052, "eval_steps_per_second": 9.813, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.07569588720798492, "learning_rate": 2.7000000000000002e-05, "loss": 0.0079, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8761061946902655, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9064327485380118, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9064327485380117, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05268806964159012, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9317073170731707, "eval_overall_precision": 0.9204819277108434, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.3006, "eval_samples_per_second": 565.527, "eval_steps_per_second": 9.98, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.6459075212478638, "learning_rate": 2.6500000000000004e-05, "loss": 0.0094, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.9282296650717704, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9150943396226415, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.8965517241379309, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8813559322033898, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05393676832318306, "eval_overall_accuracy": 0.9861916597624966, "eval_overall_f1": 0.9304029304029304, "eval_overall_precision": 0.9202898550724637, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2976, "eval_samples_per_second": 571.221, "eval_steps_per_second": 10.08, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.45708295702934265, "learning_rate": 2.6000000000000002e-05, "loss": 0.0088, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9030303030303031, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9371069182389937, "eval_ORGANIZATION_recall": 0.8713450292397661, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05946071445941925, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9366459627329193, "eval_overall_precision": 0.9425, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.2982, "eval_samples_per_second": 570.006, "eval_steps_per_second": 10.059, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.022338805720210075, "learning_rate": 2.5500000000000003e-05, "loss": 0.0071, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9194312796208531, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8981481481481481, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9032258064516129, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05394833907485008, "eval_overall_accuracy": 0.9861916597624966, "eval_overall_f1": 0.9276073619631902, "eval_overall_precision": 0.9219512195121952, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.302, "eval_samples_per_second": 562.825, "eval_steps_per_second": 9.932, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.14062269032001495, "learning_rate": 2.5e-05, "loss": 0.0079, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.9245283018867926, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8990825688073395, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9085545722713864, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9166666666666666, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05293060839176178, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9348093480934809, "eval_overall_precision": 0.9313725490196079, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2999, "eval_samples_per_second": 566.841, "eval_steps_per_second": 10.003, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.011294355615973473, "learning_rate": 2.45e-05, "loss": 0.0075, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.9201877934272301, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8909090909090909, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9144542772861357, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9226190476190477, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.054293666034936905, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.9361179361179361, "eval_overall_precision": 0.9315403422982885, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2985, "eval_samples_per_second": 569.586, "eval_steps_per_second": 10.052, "step": 4896 }, { "epoch": 52.0, "grad_norm": 0.06618872284889221, "learning_rate": 2.4e-05, "loss": 0.0078, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9201877934272301, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8909090909090909, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8979591836734693, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8953488372093024, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.054094333201646805, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.9266503667481663, "eval_overall_precision": 0.9176755447941889, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2985, "eval_samples_per_second": 569.434, "eval_steps_per_second": 10.049, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.0176588986068964, "learning_rate": 2.35e-05, "loss": 0.0075, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9282296650717704, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9150943396226415, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9117647058823529, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9171597633136095, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05433758720755577, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9371146732429099, "eval_overall_precision": 0.9359605911330049, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.3026, "eval_samples_per_second": 561.758, "eval_steps_per_second": 9.913, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.10689464211463928, "learning_rate": 2.3000000000000003e-05, "loss": 0.0081, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.9295774647887324, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9149560117302051, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9176470588235294, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05034323409199715, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9387254901960784, "eval_overall_precision": 0.9318734793187348, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.2987, "eval_samples_per_second": 569.07, "eval_steps_per_second": 10.042, "step": 5184 }, { "epoch": 55.0, "grad_norm": 0.8969389200210571, "learning_rate": 2.25e-05, "loss": 0.006, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9258160237388725, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9397590361445783, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.0562228299677372, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9445129469790382, "eval_overall_precision": 0.9433497536945813, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.3022, "eval_samples_per_second": 562.491, "eval_steps_per_second": 9.926, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.468622088432312, "learning_rate": 2.2000000000000003e-05, "loss": 0.0071, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9107142857142856, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9272727272727272, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05541842430830002, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9382716049382716, "eval_overall_precision": 0.9382716049382716, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.304, "eval_samples_per_second": 559.255, "eval_steps_per_second": 9.869, "step": 5376 }, { "epoch": 57.0, "grad_norm": 1.2091948986053467, "learning_rate": 2.15e-05, "loss": 0.0073, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.8996960486322189, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9367088607594937, "eval_ORGANIZATION_recall": 0.8654970760233918, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06389254331588745, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9339975093399752, "eval_overall_precision": 0.9422110552763819, "eval_overall_recall": 0.9259259259259259, "eval_runtime": 0.3019, "eval_samples_per_second": 563.138, "eval_steps_per_second": 9.938, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.052847035229206085, "learning_rate": 2.1e-05, "loss": 0.0048, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9226190476190477, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9393939393939394, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05634456127882004, "eval_overall_accuracy": 0.9886771610052472, "eval_overall_f1": 0.9432098765432099, "eval_overall_precision": 0.9432098765432099, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.3057, "eval_samples_per_second": 556.059, "eval_steps_per_second": 9.813, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.27387598156929016, "learning_rate": 2.05e-05, "loss": 0.0067, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9423076923076923, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9333333333333333, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9239766081871345, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9239766081871345, "eval_ORGANIZATION_recall": 0.9239766081871345, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05510587617754936, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9458128078817735, "eval_overall_precision": 0.9434889434889435, "eval_overall_recall": 0.9481481481481482, "eval_runtime": 0.3026, "eval_samples_per_second": 561.879, "eval_steps_per_second": 9.916, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.07240688055753708, "learning_rate": 2e-05, "loss": 0.0063, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9226190476190477, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9393939393939394, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05974796041846275, "eval_overall_accuracy": 0.9886771610052472, "eval_overall_f1": 0.9443757725587144, "eval_overall_precision": 0.9455445544554455, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.3046, "eval_samples_per_second": 558.087, "eval_steps_per_second": 9.849, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.16051457822322845, "learning_rate": 1.9500000000000003e-05, "loss": 0.0061, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9096209912536442, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9069767441860465, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05352047085762024, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9375764993880048, "eval_overall_precision": 0.9296116504854369, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.3019, "eval_samples_per_second": 563.117, "eval_steps_per_second": 9.937, "step": 5856 }, { "epoch": 62.0, "grad_norm": 1.3021823167800903, "learning_rate": 1.9e-05, "loss": 0.0043, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.9295774647887324, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9047619047619047, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9212121212121213, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.0623089000582695, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9346485819975339, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.3025, "eval_samples_per_second": 561.955, "eval_steps_per_second": 9.917, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.06499218940734863, "learning_rate": 1.85e-05, "loss": 0.0052, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.9377990430622011, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9245283018867925, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9090909090909091, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9117647058823529, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05624290928244591, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9359605911330049, "eval_overall_precision": 0.9336609336609336, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.3045, "eval_samples_per_second": 558.331, "eval_steps_per_second": 9.853, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.8884069323539734, "learning_rate": 1.8e-05, "loss": 0.0054, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.9289099526066351, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9074074074074074, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9085545722713864, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9166666666666666, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05896488577127457, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9334975369458127, "eval_overall_precision": 0.9312039312039312, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.3033, "eval_samples_per_second": 560.464, "eval_steps_per_second": 9.891, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.4340643286705017, "learning_rate": 1.75e-05, "loss": 0.0045, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.9238095238095239, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9065420560747663, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9164265129682997, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9034090909090909, "eval_ORGANIZATION_recall": 0.9298245614035088, "eval_PERSON_f1": 0.9809885931558935, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9772727272727273, "eval_PERSON_recall": 0.9847328244274809, "eval_loss": 0.052296627312898636, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9390243902439024, "eval_overall_precision": 0.927710843373494, "eval_overall_recall": 0.9506172839506173, "eval_runtime": 0.3033, "eval_samples_per_second": 560.5, "eval_steps_per_second": 9.891, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.015264760702848434, "learning_rate": 1.7000000000000003e-05, "loss": 0.0044, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9132947976878613, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9028571428571428, "eval_ORGANIZATION_recall": 0.9239766081871345, "eval_PERSON_f1": 0.9809885931558935, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9772727272727273, "eval_PERSON_recall": 0.9847328244274809, "eval_loss": 0.059483602643013, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9414634146341463, "eval_overall_precision": 0.9301204819277108, "eval_overall_recall": 0.9530864197530864, "eval_runtime": 0.2985, "eval_samples_per_second": 569.485, "eval_steps_per_second": 10.05, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.05331643298268318, "learning_rate": 1.65e-05, "loss": 0.0049, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.913946587537092, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.927710843373494, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9809885931558935, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9772727272727273, "eval_PERSON_recall": 0.9847328244274809, "eval_loss": 0.059692926704883575, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9421894218942188, "eval_overall_precision": 0.9387254901960784, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.2993, "eval_samples_per_second": 568.024, "eval_steps_per_second": 10.024, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.305583119392395, "learning_rate": 1.6000000000000003e-05, "loss": 0.0043, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.8988095238095237, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06636456400156021, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.934487021013597, "eval_overall_precision": 0.9356435643564357, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2979, "eval_samples_per_second": 570.695, "eval_steps_per_second": 10.071, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.006244073621928692, "learning_rate": 1.55e-05, "loss": 0.0047, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.9333333333333335, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9158878504672897, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9032258064516129, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.06099439412355423, "eval_overall_accuracy": 0.9859154929577465, "eval_overall_f1": 0.9323493234932348, "eval_overall_precision": 0.928921568627451, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2995, "eval_samples_per_second": 567.652, "eval_steps_per_second": 10.017, "step": 6624 }, { "epoch": 70.0, "grad_norm": 1.085293173789978, "learning_rate": 1.5e-05, "loss": 0.0048, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.9428571428571428, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9252336448598131, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9149560117302051, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9176470588235294, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.0641976073384285, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9421894218942188, "eval_overall_precision": 0.9387254901960784, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.2974, "eval_samples_per_second": 571.553, "eval_steps_per_second": 10.086, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.2133350968360901, "learning_rate": 1.45e-05, "loss": 0.0045, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9377990430622011, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9245283018867925, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9137931034482759, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8983050847457628, "eval_ORGANIZATION_recall": 0.9298245614035088, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05919656157493591, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9401709401709402, "eval_overall_precision": 0.9299516908212561, "eval_overall_recall": 0.9506172839506173, "eval_runtime": 0.2997, "eval_samples_per_second": 567.224, "eval_steps_per_second": 10.01, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.015173003077507019, "learning_rate": 1.4000000000000001e-05, "loss": 0.0037, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9171597633136095, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9281437125748503, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06233122944831848, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9408866995073891, "eval_overall_precision": 0.9385749385749386, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.2977, "eval_samples_per_second": 571.082, "eval_steps_per_second": 10.078, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.1929146945476532, "learning_rate": 1.3500000000000001e-05, "loss": 0.0052, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9032258064516129, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06503953784704208, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9349693251533743, "eval_overall_precision": 0.9292682926829269, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.299, "eval_samples_per_second": 568.633, "eval_steps_per_second": 10.035, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.010960678569972515, "learning_rate": 1.3000000000000001e-05, "loss": 0.0037, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9377990430622011, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9245283018867925, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9096209912536442, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9069767441860465, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05951261520385742, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9385749385749386, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.2974, "eval_samples_per_second": 571.693, "eval_steps_per_second": 10.089, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.03284645453095436, "learning_rate": 1.25e-05, "loss": 0.0028, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9161676646706587, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9386503067484663, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06371769309043884, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9419035846724351, "eval_overall_precision": 0.943069306930693, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2972, "eval_samples_per_second": 572.051, "eval_steps_per_second": 10.095, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.09497182071208954, "learning_rate": 1.2e-05, "loss": 0.0035, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9069767441860466, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9017341040462428, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.061348769813776016, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9364303178484108, "eval_overall_precision": 0.927360774818402, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.3017, "eval_samples_per_second": 563.463, "eval_steps_per_second": 9.943, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.01006114762276411, "learning_rate": 1.1500000000000002e-05, "loss": 0.0038, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.9295774647887324, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9080118694362017, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9216867469879518, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06554180383682251, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9359605911330049, "eval_overall_precision": 0.9336609336609336, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2983, "eval_samples_per_second": 569.942, "eval_steps_per_second": 10.058, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.04354839771986008, "learning_rate": 1.1000000000000001e-05, "loss": 0.0046, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9295774647887324, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9085545722713864, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9166666666666666, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06349269300699234, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9361179361179361, "eval_overall_precision": 0.9315403422982885, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2998, "eval_samples_per_second": 566.992, "eval_steps_per_second": 10.006, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.6549085974693298, "learning_rate": 1.05e-05, "loss": 0.0036, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.913946587537092, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.927710843373494, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.061547666788101196, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9395807644882861, "eval_overall_precision": 0.9384236453201971, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2979, "eval_samples_per_second": 570.628, "eval_steps_per_second": 10.07, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.557411253452301, "learning_rate": 1e-05, "loss": 0.0042, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9245283018867926, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8990825688073395, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9096209912536442, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9069767441860465, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.057892363518476486, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9351285189718482, "eval_overall_precision": 0.9271844660194175, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.2956, "eval_samples_per_second": 575.063, "eval_steps_per_second": 10.148, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.3121015131473541, "learning_rate": 9.5e-06, "loss": 0.0037, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.9289099526066351, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9074074074074074, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9090909090909091, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9117647058823529, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9731800766283524, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9769230769230769, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05845481529831886, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9348093480934809, "eval_overall_precision": 0.9313725490196079, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.3121, "eval_samples_per_second": 544.763, "eval_steps_per_second": 9.613, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.6399143934249878, "learning_rate": 9e-06, "loss": 0.0035, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9433962264150944, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9174311926605505, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9235294117647059, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9289940828402367, "eval_ORGANIZATION_recall": 0.9181286549707602, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.057273510843515396, "eval_overall_accuracy": 0.9886771610052472, "eval_overall_f1": 0.9459459459459458, "eval_overall_precision": 0.941320293398533, "eval_overall_recall": 0.9506172839506173, "eval_runtime": 0.2999, "eval_samples_per_second": 566.843, "eval_steps_per_second": 10.003, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.04775100201368332, "learning_rate": 8.500000000000002e-06, "loss": 0.0031, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9433962264150944, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9174311926605505, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9171597633136095, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9281437125748503, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.062356408685445786, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9433497536945813, "eval_overall_precision": 0.941031941031941, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.304, "eval_samples_per_second": 559.289, "eval_steps_per_second": 9.87, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.02066282369196415, "learning_rate": 8.000000000000001e-06, "loss": 0.003, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9144542772861357, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9226190476190477, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06366873532533646, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.939877300613497, "eval_overall_precision": 0.9341463414634147, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.3054, "eval_samples_per_second": 556.653, "eval_steps_per_second": 9.823, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.3127962350845337, "learning_rate": 7.5e-06, "loss": 0.0036, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.9345794392523364, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9235294117647059, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9289940828402367, "eval_ORGANIZATION_recall": 0.9181286549707602, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06440627574920654, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.943627450980392, "eval_overall_precision": 0.9367396593673966, "eval_overall_recall": 0.9506172839506173, "eval_runtime": 0.299, "eval_samples_per_second": 568.56, "eval_steps_per_second": 10.033, "step": 8160 }, { "epoch": 86.0, "grad_norm": 0.0019319866551086307, "learning_rate": 7.000000000000001e-06, "loss": 0.0035, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9345794392523364, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9107142857142856, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9272727272727272, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06361889839172363, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9372693726937269, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2988, "eval_samples_per_second": 568.964, "eval_steps_per_second": 10.041, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.006207878235727549, "learning_rate": 6.5000000000000004e-06, "loss": 0.0032, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9171597633136095, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9281437125748503, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06188651919364929, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9421894218942188, "eval_overall_precision": 0.9387254901960784, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.3023, "eval_samples_per_second": 562.349, "eval_steps_per_second": 9.924, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.0147634232416749, "learning_rate": 6e-06, "loss": 0.0029, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9127906976744187, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9075144508670521, "eval_ORGANIZATION_recall": 0.9181286549707602, "eval_PERSON_f1": 0.9731800766283524, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9769230769230769, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.06038166582584381, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.938875305623472, "eval_overall_precision": 0.9297820823244553, "eval_overall_recall": 0.9481481481481482, "eval_runtime": 0.3007, "eval_samples_per_second": 565.407, "eval_steps_per_second": 9.978, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.2083333283662796, "learning_rate": 5.500000000000001e-06, "loss": 0.0031, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9101449275362319, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9022988505747126, "eval_ORGANIZATION_recall": 0.9181286549707602, "eval_PERSON_f1": 0.9731800766283524, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9769230769230769, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.06396310776472092, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9364303178484108, "eval_overall_precision": 0.927360774818402, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.2963, "eval_samples_per_second": 573.687, "eval_steps_per_second": 10.124, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.01854861155152321, "learning_rate": 5e-06, "loss": 0.0028, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9345794392523364, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.913946587537092, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.927710843373494, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06590648740530014, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.939729397293973, "eval_overall_precision": 0.9362745098039216, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.2978, "eval_samples_per_second": 570.899, "eval_steps_per_second": 10.075, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.06623278558254242, "learning_rate": 4.5e-06, "loss": 0.0028, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9208211143695014, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9235294117647059, "eval_ORGANIZATION_recall": 0.9181286549707602, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06273588538169861, "eval_overall_accuracy": 0.9886771610052472, "eval_overall_f1": 0.9434889434889435, "eval_overall_precision": 0.9388753056234719, "eval_overall_recall": 0.9481481481481482, "eval_runtime": 0.351, "eval_samples_per_second": 484.303, "eval_steps_per_second": 8.547, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.02434733510017395, "learning_rate": 4.000000000000001e-06, "loss": 0.0032, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9171597633136095, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9281437125748503, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06576044112443924, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9421894218942188, "eval_overall_precision": 0.9387254901960784, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.2994, "eval_samples_per_second": 567.823, "eval_steps_per_second": 10.02, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.6385473012924194, "learning_rate": 3.5000000000000004e-06, "loss": 0.0023, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9433962264150944, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9174311926605505, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9176470588235294, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9230769230769231, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06361746042966843, "eval_overall_accuracy": 0.9884009942004971, "eval_overall_f1": 0.9434889434889435, "eval_overall_precision": 0.9388753056234719, "eval_overall_recall": 0.9481481481481482, "eval_runtime": 0.3009, "eval_samples_per_second": 565.033, "eval_steps_per_second": 9.971, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.06679230183362961, "learning_rate": 3e-06, "loss": 0.0024, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.9433962264150944, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9174311926605505, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9226190476190477, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9393939393939394, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06507156044244766, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9456790123456791, "eval_overall_precision": 0.945679012345679, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.2976, "eval_samples_per_second": 571.157, "eval_steps_per_second": 10.079, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.035074394196271896, "learning_rate": 2.5e-06, "loss": 0.0029, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9090909090909091, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9117647058823529, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06322455406188965, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9385749385749386, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.2983, "eval_samples_per_second": 569.901, "eval_steps_per_second": 10.057, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.34420645236968994, "learning_rate": 2.0000000000000003e-06, "loss": 0.0021, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9176470588235294, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9230769230769231, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06341253966093063, "eval_overall_accuracy": 0.9884009942004971, "eval_overall_f1": 0.9423312883435584, "eval_overall_precision": 0.9365853658536586, "eval_overall_recall": 0.9481481481481482, "eval_runtime": 0.2987, "eval_samples_per_second": 569.166, "eval_steps_per_second": 10.044, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.01225065067410469, "learning_rate": 1.5e-06, "loss": 0.0027, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9226190476190477, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9393939393939394, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06498675793409348, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9445129469790382, "eval_overall_precision": 0.9433497536945813, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.2988, "eval_samples_per_second": 568.927, "eval_steps_per_second": 10.04, "step": 9312 }, { "epoch": 98.0, "grad_norm": 0.060756802558898926, "learning_rate": 1.0000000000000002e-06, "loss": 0.0019, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9226190476190477, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9393939393939394, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.0640244409441948, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9445129469790382, "eval_overall_precision": 0.9433497536945813, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.299, "eval_samples_per_second": 568.58, "eval_steps_per_second": 10.034, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.023820942267775536, "learning_rate": 5.000000000000001e-07, "loss": 0.0026, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9226190476190477, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9393939393939394, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06459710747003555, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9445129469790382, "eval_overall_precision": 0.9433497536945813, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.3005, "eval_samples_per_second": 565.803, "eval_steps_per_second": 9.985, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.4159657955169678, "learning_rate": 0.0, "loss": 0.0027, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9171597633136095, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9281437125748503, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.06436590105295181, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9421894218942188, "eval_overall_precision": 0.9387254901960784, "eval_overall_recall": 0.945679012345679, "eval_runtime": 0.2969, "eval_samples_per_second": 572.522, "eval_steps_per_second": 10.103, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 4381316244449244.0, "train_loss": 0.03098656691610813, "train_runtime": 869.7405, "train_samples_per_second": 176.03, "train_steps_per_second": 11.038 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 4381316244449244.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }