{ "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": 1.0839929580688477, "learning_rate": 4.9500000000000004e-05, "loss": 0.8467, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.0, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.0, "eval_LOCATION_recall": 0.0, "eval_ORGANIZATION_f1": 0.23076923076923078, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.24375, "eval_ORGANIZATION_recall": 0.21910112359550563, "eval_PERSON_f1": 0.3233082706766917, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.3115942028985507, "eval_PERSON_recall": 0.3359375, "eval_loss": 0.3800315260887146, "eval_overall_accuracy": 0.8699055330634278, "eval_overall_f1": 0.23699421965317918, "eval_overall_precision": 0.2733333333333333, "eval_overall_recall": 0.20918367346938777, "eval_runtime": 0.2945, "eval_samples_per_second": 577.299, "eval_steps_per_second": 10.188, "step": 96 }, { "epoch": 2.0, "grad_norm": 0.755494236946106, "learning_rate": 4.9e-05, "loss": 0.3716, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.38823529411764707, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.39285714285714285, "eval_LOCATION_recall": 0.38372093023255816, "eval_ORGANIZATION_f1": 0.5763688760806918, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.591715976331361, "eval_ORGANIZATION_recall": 0.5617977528089888, "eval_PERSON_f1": 0.7132867132867134, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.6455696202531646, "eval_PERSON_recall": 0.796875, "eval_loss": 0.24234628677368164, "eval_overall_accuracy": 0.9309041835357624, "eval_overall_f1": 0.585305105853051, "eval_overall_precision": 0.5717761557177615, "eval_overall_recall": 0.5994897959183674, "eval_runtime": 0.3009, "eval_samples_per_second": 564.972, "eval_steps_per_second": 9.97, "step": 192 }, { "epoch": 3.0, "grad_norm": 0.8059985637664795, "learning_rate": 4.85e-05, "loss": 0.2184, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.736842105263158, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.7411764705882353, "eval_LOCATION_recall": 0.7325581395348837, "eval_ORGANIZATION_f1": 0.7479674796747966, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.7225130890052356, "eval_ORGANIZATION_recall": 0.7752808988764045, "eval_PERSON_f1": 0.9473684210526315, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9130434782608695, "eval_PERSON_recall": 0.984375, "eval_loss": 0.12176795303821564, "eval_overall_accuracy": 0.9649122807017544, "eval_overall_f1": 0.8114143920595533, "eval_overall_precision": 0.7898550724637681, "eval_overall_recall": 0.8341836734693877, "eval_runtime": 0.2989, "eval_samples_per_second": 568.667, "eval_steps_per_second": 10.035, "step": 288 }, { "epoch": 4.0, "grad_norm": 0.6826747059822083, "learning_rate": 4.8e-05, "loss": 0.1455, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.7564766839378237, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.6822429906542056, "eval_LOCATION_recall": 0.8488372093023255, "eval_ORGANIZATION_f1": 0.7887323943661972, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.7909604519774012, "eval_ORGANIZATION_recall": 0.7865168539325843, "eval_PERSON_f1": 0.9505703422053232, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9259259259259259, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.0960189700126648, "eval_overall_accuracy": 0.9713900134952766, "eval_overall_f1": 0.8335388409371147, "eval_overall_precision": 0.8066825775656324, "eval_overall_recall": 0.8622448979591837, "eval_runtime": 0.3022, "eval_samples_per_second": 562.561, "eval_steps_per_second": 9.928, "step": 384 }, { "epoch": 5.0, "grad_norm": 0.6259937286376953, "learning_rate": 4.75e-05, "loss": 0.1147, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.839779005524862, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8, "eval_LOCATION_recall": 0.8837209302325582, "eval_ORGANIZATION_f1": 0.8784530386740331, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.8641304347826086, "eval_ORGANIZATION_recall": 0.8932584269662921, "eval_PERSON_f1": 0.9578544061302682, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9398496240601504, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.07682560384273529, "eval_overall_accuracy": 0.9762483130904184, "eval_overall_f1": 0.8955223880597014, "eval_overall_precision": 0.8737864077669902, "eval_overall_recall": 0.9183673469387755, "eval_runtime": 0.3033, "eval_samples_per_second": 560.446, "eval_steps_per_second": 9.89, "step": 480 }, { "epoch": 6.0, "grad_norm": 0.8320963382720947, "learning_rate": 4.7e-05, "loss": 0.1009, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.7575757575757577, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.6696428571428571, "eval_LOCATION_recall": 0.872093023255814, "eval_ORGANIZATION_f1": 0.8520710059171598, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9, "eval_ORGANIZATION_recall": 0.8089887640449438, "eval_PERSON_f1": 0.9689922480620154, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9615384615384616, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.07274217903614044, "eval_overall_accuracy": 0.9775978407557355, "eval_overall_f1": 0.8664987405541562, "eval_overall_precision": 0.8557213930348259, "eval_overall_recall": 0.8775510204081632, "eval_runtime": 0.2981, "eval_samples_per_second": 570.228, "eval_steps_per_second": 10.063, "step": 576 }, { "epoch": 7.0, "grad_norm": 0.6125458478927612, "learning_rate": 4.6500000000000005e-05, "loss": 0.0893, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.8524590163934426, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8041237113402062, "eval_LOCATION_recall": 0.9069767441860465, "eval_ORGANIZATION_f1": 0.8870523415977961, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.8702702702702703, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.05971812084317207, "eval_overall_accuracy": 0.9813765182186235, "eval_overall_f1": 0.9052369077306733, "eval_overall_precision": 0.8853658536585366, "eval_overall_recall": 0.9260204081632653, "eval_runtime": 0.2962, "eval_samples_per_second": 573.963, "eval_steps_per_second": 10.129, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.2061917930841446, "learning_rate": 4.600000000000001e-05, "loss": 0.082, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.875, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8555555555555555, "eval_LOCATION_recall": 0.8953488372093024, "eval_ORGANIZATION_f1": 0.8870523415977961, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.8702702702702703, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9609375, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9609375, "eval_loss": 0.04958827793598175, "eval_overall_accuracy": 0.9835357624831309, "eval_overall_f1": 0.9081761006289308, "eval_overall_precision": 0.8957816377171216, "eval_overall_recall": 0.9209183673469388, "eval_runtime": 0.2976, "eval_samples_per_second": 571.277, "eval_steps_per_second": 10.081, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.37945252656936646, "learning_rate": 4.55e-05, "loss": 0.0775, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.8449197860962566, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.7821782178217822, "eval_LOCATION_recall": 0.9186046511627907, "eval_ORGANIZATION_f1": 0.8850574712643678, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.8651685393258427, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05077804625034332, "eval_overall_accuracy": 0.9811066126855601, "eval_overall_f1": 0.9051833122629582, "eval_overall_precision": 0.8972431077694235, "eval_overall_recall": 0.9132653061224489, "eval_runtime": 0.3, "eval_samples_per_second": 566.755, "eval_steps_per_second": 10.002, "step": 864 }, { "epoch": 10.0, "grad_norm": 0.8556278944015503, "learning_rate": 4.5e-05, "loss": 0.0696, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.8481675392670157, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.7714285714285715, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.8971428571428571, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9127906976744186, "eval_ORGANIZATION_recall": 0.8820224719101124, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.0515359602868557, "eval_overall_accuracy": 0.9819163292847504, "eval_overall_f1": 0.9120603015075376, "eval_overall_precision": 0.8985148514851485, "eval_overall_recall": 0.9260204081632653, "eval_runtime": 0.3007, "eval_samples_per_second": 565.294, "eval_steps_per_second": 9.976, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.9886091947555542, "learning_rate": 4.4500000000000004e-05, "loss": 0.0635, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.88268156424581, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8494623655913979, "eval_LOCATION_recall": 0.9186046511627907, "eval_ORGANIZATION_f1": 0.935933147632312, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9281767955801105, "eval_ORGANIZATION_recall": 0.9438202247191011, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.043357785791158676, "eval_overall_accuracy": 0.9854251012145749, "eval_overall_f1": 0.9345088161209069, "eval_overall_precision": 0.9228855721393034, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2996, "eval_samples_per_second": 567.344, "eval_steps_per_second": 10.012, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.3404422700405121, "learning_rate": 4.4000000000000006e-05, "loss": 0.0604, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.9152542372881357, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8901098901098901, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9175824175824175, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.8978494623655914, "eval_ORGANIZATION_recall": 0.9382022471910112, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.042961485683918, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9371859296482412, "eval_overall_precision": 0.9232673267326733, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2993, "eval_samples_per_second": 567.954, "eval_steps_per_second": 10.023, "step": 1152 }, { "epoch": 13.0, "grad_norm": 1.0723609924316406, "learning_rate": 4.35e-05, "loss": 0.0557, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.9080459770114941, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8977272727272727, "eval_LOCATION_recall": 0.9186046511627907, "eval_ORGANIZATION_f1": 0.9175824175824175, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.8978494623655914, "eval_ORGANIZATION_recall": 0.9382022471910112, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.041559427976608276, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.9319899244332494, "eval_overall_precision": 0.9203980099502488, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2985, "eval_samples_per_second": 569.598, "eval_steps_per_second": 10.052, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.3902477025985718, "learning_rate": 4.3e-05, "loss": 0.051, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.9132947976878613, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9080459770114943, "eval_LOCATION_recall": 0.9186046511627907, "eval_ORGANIZATION_f1": 0.9157303370786517, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9157303370786517, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04430091381072998, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.9362244897959183, "eval_overall_precision": 0.9362244897959183, "eval_overall_recall": 0.9362244897959183, "eval_runtime": 0.2993, "eval_samples_per_second": 567.929, "eval_steps_per_second": 10.022, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.8345674872398376, "learning_rate": 4.25e-05, "loss": 0.0492, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.8977272727272728, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8777777777777778, "eval_LOCATION_recall": 0.9186046511627907, "eval_ORGANIZATION_f1": 0.9126760563380282, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9152542372881356, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9725490196078432, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9763779527559056, "eval_PERSON_recall": 0.96875, "eval_loss": 0.04240505024790764, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9287531806615775, "eval_overall_precision": 0.9263959390862944, "eval_overall_recall": 0.9311224489795918, "eval_runtime": 0.2999, "eval_samples_per_second": 566.824, "eval_steps_per_second": 10.003, "step": 1440 }, { "epoch": 16.0, "grad_norm": 0.7208335995674133, "learning_rate": 4.2e-05, "loss": 0.0463, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.9265536723163842, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9010989010989011, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9307479224376731, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9180327868852459, "eval_ORGANIZATION_recall": 0.9438202247191011, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04386291652917862, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9457755359394704, "eval_overall_precision": 0.9351620947630923, "eval_overall_recall": 0.9566326530612245, "eval_runtime": 0.2981, "eval_samples_per_second": 570.24, "eval_steps_per_second": 10.063, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.3200893998146057, "learning_rate": 4.15e-05, "loss": 0.0417, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.9204545454545455, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9362880886426592, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9234972677595629, "eval_ORGANIZATION_recall": 0.949438202247191, "eval_PERSON_f1": 0.9727626459143969, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9689922480620154, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04438333958387375, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9445843828715365, "eval_overall_precision": 0.9328358208955224, "eval_overall_recall": 0.9566326530612245, "eval_runtime": 0.3012, "eval_samples_per_second": 564.398, "eval_steps_per_second": 9.96, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.3097224831581116, "learning_rate": 4.1e-05, "loss": 0.0407, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.9239766081871345, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9294117647058824, "eval_LOCATION_recall": 0.9186046511627907, "eval_ORGANIZATION_f1": 0.9385474860335196, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9333333333333333, "eval_ORGANIZATION_recall": 0.9438202247191011, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.0414746068418026, "eval_overall_accuracy": 0.9883940620782726, "eval_overall_f1": 0.9489795918367347, "eval_overall_precision": 0.9489795918367347, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2986, "eval_samples_per_second": 569.248, "eval_steps_per_second": 10.046, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.6831247806549072, "learning_rate": 4.05e-05, "loss": 0.0413, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.9132947976878613, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9080459770114943, "eval_LOCATION_recall": 0.9186046511627907, "eval_ORGANIZATION_f1": 0.9497206703910613, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9444444444444444, "eval_ORGANIZATION_recall": 0.9550561797752809, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04294687882065773, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9516539440203563, "eval_overall_precision": 0.949238578680203, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.2977, "eval_samples_per_second": 571.08, "eval_steps_per_second": 10.078, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.28463003039360046, "learning_rate": 4e-05, "loss": 0.0368, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9187675070028011, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9162011173184358, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9727626459143969, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9689922480620154, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05581528693437576, "eval_overall_accuracy": 0.9838056680161943, "eval_overall_f1": 0.9333333333333335, "eval_overall_precision": 0.9205955334987593, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2974, "eval_samples_per_second": 571.576, "eval_steps_per_second": 10.087, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.36661648750305176, "learning_rate": 3.9500000000000005e-05, "loss": 0.0361, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.9204545454545455, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9269662921348315, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9269662921348315, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9725490196078432, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9763779527559056, "eval_PERSON_recall": 0.96875, "eval_loss": 0.04730469360947609, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9402795425667091, "eval_overall_precision": 0.9367088607594937, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2981, "eval_samples_per_second": 570.287, "eval_steps_per_second": 10.064, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.545528769493103, "learning_rate": 3.9000000000000006e-05, "loss": 0.0355, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.8743169398907104, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8247422680412371, "eval_LOCATION_recall": 0.9302325581395349, "eval_ORGANIZATION_f1": 0.9171270718232043, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9021739130434783, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9609375, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9609375, "eval_loss": 0.05565020069479942, "eval_overall_accuracy": 0.9835357624831309, "eval_overall_f1": 0.9213483146067415, "eval_overall_precision": 0.902200488997555, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.3002, "eval_samples_per_second": 566.326, "eval_steps_per_second": 9.994, "step": 2112 }, { "epoch": 23.0, "grad_norm": 0.3839501440525055, "learning_rate": 3.85e-05, "loss": 0.0314, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.9152542372881357, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8901098901098901, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9411764705882353, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9385474860335196, "eval_ORGANIZATION_recall": 0.9438202247191011, "eval_PERSON_f1": 0.9689922480620154, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9615384615384616, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04751402512192726, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9444444444444445, "eval_overall_precision": 0.935, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.3015, "eval_samples_per_second": 563.806, "eval_steps_per_second": 9.95, "step": 2208 }, { "epoch": 24.0, "grad_norm": 0.2973721921443939, "learning_rate": 3.8e-05, "loss": 0.0308, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.8972972972972972, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8383838383838383, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9195402298850575, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9411764705882353, "eval_ORGANIZATION_recall": 0.898876404494382, "eval_PERSON_f1": 0.9649805447470817, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9612403100775194, "eval_PERSON_recall": 0.96875, "eval_loss": 0.05555843934416771, "eval_overall_accuracy": 0.9851551956815114, "eval_overall_f1": 0.929113924050633, "eval_overall_precision": 0.9221105527638191, "eval_overall_recall": 0.9362244897959183, "eval_runtime": 0.3025, "eval_samples_per_second": 561.908, "eval_steps_per_second": 9.916, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.34367889165878296, "learning_rate": 3.7500000000000003e-05, "loss": 0.0303, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.9371428571428573, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9213483146067416, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9291784702549575, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9371428571428572, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.05049148574471474, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9438775510204082, "eval_overall_precision": 0.9438775510204082, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.3, "eval_samples_per_second": 566.687, "eval_steps_per_second": 10.0, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.5910239219665527, "learning_rate": 3.7e-05, "loss": 0.0288, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9318181818181819, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9111111111111111, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9157303370786517, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9157303370786517, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04861859232187271, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2974, "eval_samples_per_second": 571.593, "eval_steps_per_second": 10.087, "step": 2496 }, { "epoch": 27.0, "grad_norm": 0.20005720853805542, "learning_rate": 3.65e-05, "loss": 0.0276, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9277777777777777, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9175824175824175, "eval_ORGANIZATION_recall": 0.9382022471910112, "eval_PERSON_f1": 0.9689922480620154, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9615384615384616, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04790767282247543, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9385194479297364, "eval_overall_precision": 0.9234567901234568, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.297, "eval_samples_per_second": 572.323, "eval_steps_per_second": 10.1, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.9156953692436218, "learning_rate": 3.6e-05, "loss": 0.0264, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9411764705882352, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9523809523809523, "eval_LOCATION_recall": 0.9302325581395349, "eval_ORGANIZATION_f1": 0.9205479452054793, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.8983957219251337, "eval_ORGANIZATION_recall": 0.9438202247191011, "eval_PERSON_f1": 0.9571984435797667, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9609375, "eval_loss": 0.04682742431759834, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9368686868686869, "eval_overall_precision": 0.9275, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2987, "eval_samples_per_second": 569.044, "eval_steps_per_second": 10.042, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.8349704742431641, "learning_rate": 3.55e-05, "loss": 0.0267, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9183098591549295, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9209039548022598, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.055273257195949554, "eval_overall_accuracy": 0.9848852901484481, "eval_overall_f1": 0.9343434343434343, "eval_overall_precision": 0.925, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.3063, "eval_samples_per_second": 555.038, "eval_steps_per_second": 9.795, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.16796889901161194, "learning_rate": 3.5e-05, "loss": 0.0258, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.9142857142857143, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.898876404494382, "eval_LOCATION_recall": 0.9302325581395349, "eval_ORGANIZATION_f1": 0.934065934065934, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9139784946236559, "eval_ORGANIZATION_recall": 0.9550561797752809, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04855236038565636, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9433962264150944, "eval_overall_precision": 0.9305210918114144, "eval_overall_recall": 0.9566326530612245, "eval_runtime": 0.3006, "eval_samples_per_second": 565.457, "eval_steps_per_second": 9.979, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.3600468635559082, "learning_rate": 3.45e-05, "loss": 0.0252, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.9204545454545455, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9689922480620154, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9615384615384616, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05072575807571411, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.9389312977099236, "eval_overall_precision": 0.9365482233502538, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2984, "eval_samples_per_second": 569.615, "eval_steps_per_second": 10.052, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.7236989140510559, "learning_rate": 3.4000000000000007e-05, "loss": 0.0248, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9318181818181819, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9111111111111111, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9101123595505618, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9101123595505618, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9649805447470817, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9612403100775194, "eval_PERSON_recall": 0.96875, "eval_loss": 0.04983556270599365, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9328263624841572, "eval_overall_precision": 0.9269521410579346, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.3007, "eval_samples_per_second": 565.425, "eval_steps_per_second": 9.978, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.3026004433631897, "learning_rate": 3.35e-05, "loss": 0.0233, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.942528735632184, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9318181818181818, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9325842696629213, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9325842696629213, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05156313255429268, "eval_overall_accuracy": 0.9881241565452091, "eval_overall_f1": 0.9491094147582698, "eval_overall_precision": 0.9467005076142132, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.3013, "eval_samples_per_second": 564.287, "eval_steps_per_second": 9.958, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.5758619904518127, "learning_rate": 3.3e-05, "loss": 0.0243, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9382022471910112, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9382022471910112, "eval_ORGANIZATION_recall": 0.9382022471910112, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.054125815629959106, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9444444444444445, "eval_overall_precision": 0.935, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.3009, "eval_samples_per_second": 564.913, "eval_steps_per_second": 9.969, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.5928752422332764, "learning_rate": 3.2500000000000004e-05, "loss": 0.0199, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.9318181818181819, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9111111111111111, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9243697478991597, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9217877094972067, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05149749293923378, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9429657794676806, "eval_overall_precision": 0.9370277078085643, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.3012, "eval_samples_per_second": 564.434, "eval_steps_per_second": 9.961, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.46633267402648926, "learning_rate": 3.2000000000000005e-05, "loss": 0.0217, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.8961748633879781, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.845360824742268, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9022988505747127, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9235294117647059, "eval_ORGANIZATION_recall": 0.8820224719101124, "eval_PERSON_f1": 0.9647058823529412, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9609375, "eval_loss": 0.06958433985710144, "eval_overall_accuracy": 0.982995951417004, "eval_overall_f1": 0.9211195928753181, "eval_overall_precision": 0.9187817258883249, "eval_overall_recall": 0.923469387755102, "eval_runtime": 0.2981, "eval_samples_per_second": 570.214, "eval_steps_per_second": 10.063, "step": 3456 }, { "epoch": 37.0, "grad_norm": 0.4119836986064911, "learning_rate": 3.15e-05, "loss": 0.0215, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.9090909090909092, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.9302325581395349, "eval_ORGANIZATION_f1": 0.9106145251396648, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9055555555555556, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05974140763282776, "eval_overall_accuracy": 0.9856950067476383, "eval_overall_f1": 0.9316455696202531, "eval_overall_precision": 0.9246231155778895, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.3015, "eval_samples_per_second": 563.791, "eval_steps_per_second": 9.949, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.1353997439146042, "learning_rate": 3.1e-05, "loss": 0.0194, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9111111111111112, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8723404255319149, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9183098591549295, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9209039548022598, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9725490196078432, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9763779527559056, "eval_PERSON_recall": 0.96875, "eval_loss": 0.05276887118816376, "eval_overall_accuracy": 0.9856950067476383, "eval_overall_f1": 0.9341772151898734, "eval_overall_precision": 0.9271356783919598, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2984, "eval_samples_per_second": 569.644, "eval_steps_per_second": 10.053, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.31559664011001587, "learning_rate": 3.05e-05, "loss": 0.0182, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.92090395480226, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9261363636363636, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9727626459143969, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9689922480620154, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05161591246724129, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9378960709759189, "eval_overall_precision": 0.9319899244332494, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2971, "eval_samples_per_second": 572.14, "eval_steps_per_second": 10.097, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.6130654215812683, "learning_rate": 3e-05, "loss": 0.0187, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.9273743016759777, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8924731182795699, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05681144446134567, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9428208386277002, "eval_overall_precision": 0.9392405063291139, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.3002, "eval_samples_per_second": 566.35, "eval_steps_per_second": 9.994, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.11302889138460159, "learning_rate": 2.95e-05, "loss": 0.0184, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.9050279329608939, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8709677419354839, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9111747851002865, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9298245614035088, "eval_ORGANIZATION_recall": 0.8932584269662921, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06318608671426773, "eval_overall_accuracy": 0.9843454790823212, "eval_overall_f1": 0.9311224489795918, "eval_overall_precision": 0.9311224489795918, "eval_overall_recall": 0.9311224489795918, "eval_runtime": 0.2988, "eval_samples_per_second": 568.92, "eval_steps_per_second": 10.04, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.43945929408073425, "learning_rate": 2.9e-05, "loss": 0.0175, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9126760563380282, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9152542372881356, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9689922480620154, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9615384615384616, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.055987853556871414, "eval_overall_accuracy": 0.9851551956815114, "eval_overall_f1": 0.9318181818181819, "eval_overall_precision": 0.9225, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2996, "eval_samples_per_second": 567.494, "eval_steps_per_second": 10.015, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.7528828382492065, "learning_rate": 2.8499999999999998e-05, "loss": 0.0177, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9204545454545455, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9157303370786517, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9157303370786517, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9727626459143969, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9689922480620154, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05622381344437599, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.935361216730038, "eval_overall_precision": 0.929471032745592, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2982, "eval_samples_per_second": 570.136, "eval_steps_per_second": 10.061, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.342618465423584, "learning_rate": 2.8000000000000003e-05, "loss": 0.0181, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9318181818181819, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9111111111111111, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9157303370786517, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9157303370786517, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.050536252558231354, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.9365482233502538, "eval_overall_precision": 0.9318181818181818, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2982, "eval_samples_per_second": 570.013, "eval_steps_per_second": 10.059, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.4087114632129669, "learning_rate": 2.7500000000000004e-05, "loss": 0.0172, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.9111111111111112, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8723404255319149, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.905982905982906, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9190751445086706, "eval_ORGANIZATION_recall": 0.8932584269662921, "eval_PERSON_f1": 0.9571984435797667, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9609375, "eval_loss": 0.06356929987668991, "eval_overall_accuracy": 0.9854251012145749, "eval_overall_f1": 0.9238578680203046, "eval_overall_precision": 0.9191919191919192, "eval_overall_recall": 0.9285714285714286, "eval_runtime": 0.2963, "eval_samples_per_second": 573.707, "eval_steps_per_second": 10.124, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.832419753074646, "learning_rate": 2.7000000000000002e-05, "loss": 0.0166, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.9171270718232045, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8736842105263158, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9028571428571427, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9186046511627907, "eval_ORGANIZATION_recall": 0.8876404494382022, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06471683084964752, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.9301143583227446, "eval_overall_precision": 0.9265822784810127, "eval_overall_recall": 0.9336734693877551, "eval_runtime": 0.2962, "eval_samples_per_second": 573.994, "eval_steps_per_second": 10.129, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.4743807315826416, "learning_rate": 2.6500000000000004e-05, "loss": 0.017, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.9152542372881357, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8901098901098901, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9187675070028011, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9162011173184358, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05505603179335594, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9367088607594937, "eval_overall_precision": 0.9296482412060302, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2997, "eval_samples_per_second": 567.195, "eval_steps_per_second": 10.009, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.09081308543682098, "learning_rate": 2.6000000000000002e-05, "loss": 0.0147, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9111111111111112, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8723404255319149, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9096045197740114, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9147727272727273, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.057248689234256744, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9316455696202531, "eval_overall_precision": 0.9246231155778895, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.2958, "eval_samples_per_second": 574.809, "eval_steps_per_second": 10.144, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.29989704489707947, "learning_rate": 2.5500000000000003e-05, "loss": 0.0138, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9199999999999999, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.936046511627907, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05268542468547821, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.937579617834395, "eval_overall_precision": 0.9363867684478372, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.2963, "eval_samples_per_second": 573.646, "eval_steps_per_second": 10.123, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.2130219042301178, "learning_rate": 2.5e-05, "loss": 0.0154, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.9418604651162791, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9418604651162791, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9307479224376731, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9180327868852459, "eval_ORGANIZATION_recall": 0.9438202247191011, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06127961724996567, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9480354879594423, "eval_overall_precision": 0.9420654911838791, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.2947, "eval_samples_per_second": 576.898, "eval_steps_per_second": 10.181, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.063932403922081, "learning_rate": 2.45e-05, "loss": 0.0151, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9287749287749287, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9421965317919075, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.061540693044662476, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.94147582697201, "eval_overall_precision": 0.9390862944162437, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.294, "eval_samples_per_second": 578.322, "eval_steps_per_second": 10.206, "step": 4896 }, { "epoch": 52.0, "grad_norm": 0.16586625576019287, "learning_rate": 2.4e-05, "loss": 0.0144, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9291784702549575, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9371428571428572, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.054826878011226654, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9402795425667091, "eval_overall_precision": 0.9367088607594937, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2951, "eval_samples_per_second": 576.102, "eval_steps_per_second": 10.167, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.5482709407806396, "learning_rate": 2.35e-05, "loss": 0.0145, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9132947976878613, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9080459770114943, "eval_LOCATION_recall": 0.9186046511627907, "eval_ORGANIZATION_f1": 0.9269662921348315, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9269662921348315, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06688716262578964, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.940127388535032, "eval_overall_precision": 0.9389312977099237, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2937, "eval_samples_per_second": 578.915, "eval_steps_per_second": 10.216, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.23728640377521515, "learning_rate": 2.3000000000000003e-05, "loss": 0.0132, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.9195402298850575, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.9302325581395349, "eval_ORGANIZATION_f1": 0.9131652661064427, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9106145251396648, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.07014357298612595, "eval_overall_accuracy": 0.9851551956815114, "eval_overall_f1": 0.9326556543837357, "eval_overall_precision": 0.9291139240506329, "eval_overall_recall": 0.9362244897959183, "eval_runtime": 0.2955, "eval_samples_per_second": 575.218, "eval_steps_per_second": 10.151, "step": 5184 }, { "epoch": 55.0, "grad_norm": 0.5555434823036194, "learning_rate": 2.25e-05, "loss": 0.0133, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.9204545454545455, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9257142857142857, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9418604651162791, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.059308670461177826, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9411764705882353, "eval_overall_precision": 0.9435897435897436, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.2961, "eval_samples_per_second": 574.105, "eval_steps_per_second": 10.131, "step": 5280 }, { "epoch": 56.0, "grad_norm": 1.7257238626480103, "learning_rate": 2.2000000000000003e-05, "loss": 0.0126, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9265536723163842, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9318181818181818, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05880829319357872, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9404309252217997, "eval_overall_precision": 0.9345088161209067, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2987, "eval_samples_per_second": 569.153, "eval_steps_per_second": 10.044, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.03070933371782303, "learning_rate": 2.15e-05, "loss": 0.013, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.9171270718232045, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8736842105263158, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9291784702549575, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9371428571428572, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.058942895382642746, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9417721518987343, "eval_overall_precision": 0.9346733668341709, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2951, "eval_samples_per_second": 576.154, "eval_steps_per_second": 10.167, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.5987908840179443, "learning_rate": 2.1e-05, "loss": 0.0126, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9318181818181818, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9425287356321839, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06152055785059929, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9440203562340966, "eval_overall_precision": 0.9416243654822335, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2948, "eval_samples_per_second": 576.685, "eval_steps_per_second": 10.177, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.14170171320438385, "learning_rate": 2.05e-05, "loss": 0.0113, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9273743016759777, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8924731182795699, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.059575069695711136, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9402795425667091, "eval_overall_precision": 0.9367088607594937, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2959, "eval_samples_per_second": 574.543, "eval_steps_per_second": 10.139, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.5523650646209717, "learning_rate": 2e-05, "loss": 0.0125, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.9273743016759777, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8924731182795699, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9375000000000001, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9482758620689655, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.05963626131415367, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9453621346886911, "eval_overall_precision": 0.9417721518987342, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2966, "eval_samples_per_second": 573.16, "eval_steps_per_second": 10.115, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.36497360467910767, "learning_rate": 1.9500000000000003e-05, "loss": 0.0117, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.9257142857142857, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9101123595505618, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9322033898305084, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9375, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06436024606227875, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9426751592356687, "eval_overall_precision": 0.9414758269720102, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2935, "eval_samples_per_second": 579.297, "eval_steps_per_second": 10.223, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.526648223400116, "learning_rate": 1.9e-05, "loss": 0.0113, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9152542372881356, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9204545454545454, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06210213899612427, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.9328263624841572, "eval_overall_precision": 0.9269521410579346, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.2962, "eval_samples_per_second": 573.927, "eval_steps_per_second": 10.128, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.22596602141857147, "learning_rate": 1.85e-05, "loss": 0.0115, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9204545454545454, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9310344827586207, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.0633787289261818, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9377382465057178, "eval_overall_precision": 0.9341772151898734, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2978, "eval_samples_per_second": 570.869, "eval_steps_per_second": 10.074, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.016681797802448273, "learning_rate": 1.8e-05, "loss": 0.0105, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9178470254957507, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9257142857142857, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06876453012228012, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.9365482233502538, "eval_overall_precision": 0.9318181818181818, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2951, "eval_samples_per_second": 576.125, "eval_steps_per_second": 10.167, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.029853196814656258, "learning_rate": 1.75e-05, "loss": 0.0115, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9243697478991597, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9217877094972067, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.059058159589767456, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9405815423514539, "eval_overall_precision": 0.9323308270676691, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.296, "eval_samples_per_second": 574.32, "eval_steps_per_second": 10.135, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.4149913191795349, "learning_rate": 1.7000000000000003e-05, "loss": 0.0081, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9318181818181818, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9425287356321839, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06314166635274887, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.94147582697201, "eval_overall_precision": 0.9390862944162437, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.295, "eval_samples_per_second": 576.188, "eval_steps_per_second": 10.168, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.06600440293550491, "learning_rate": 1.65e-05, "loss": 0.0091, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9204545454545454, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9310344827586207, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06787087023258209, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.9377382465057178, "eval_overall_precision": 0.9341772151898734, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2948, "eval_samples_per_second": 576.601, "eval_steps_per_second": 10.175, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.10460586100816727, "learning_rate": 1.6000000000000003e-05, "loss": 0.0096, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9265536723163842, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9010989010989011, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.92090395480226, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9261363636363636, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06838060915470123, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.9377382465057178, "eval_overall_precision": 0.9341772151898734, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2971, "eval_samples_per_second": 572.269, "eval_steps_per_second": 10.099, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.7347772717475891, "learning_rate": 1.55e-05, "loss": 0.0098, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9235127478753541, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9314285714285714, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06859349459409714, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.9402795425667091, "eval_overall_precision": 0.9367088607594937, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2951, "eval_samples_per_second": 576.066, "eval_steps_per_second": 10.166, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.21133266389369965, "learning_rate": 1.5e-05, "loss": 0.0081, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9204545454545454, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9310344827586207, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06589078158140182, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9377382465057178, "eval_overall_precision": 0.9341772151898734, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2955, "eval_samples_per_second": 575.219, "eval_steps_per_second": 10.151, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.43516725301742554, "learning_rate": 1.45e-05, "loss": 0.0091, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9405099150141643, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9485714285714286, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06766483932733536, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9479034307496824, "eval_overall_precision": 0.9443037974683545, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.296, "eval_samples_per_second": 574.269, "eval_steps_per_second": 10.134, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.06414016336202621, "learning_rate": 1.4000000000000001e-05, "loss": 0.0087, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9283667621776504, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9473684210526315, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06591153889894485, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9425287356321839, "eval_overall_precision": 0.9437340153452686, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2954, "eval_samples_per_second": 575.575, "eval_steps_per_second": 10.157, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.3230314552783966, "learning_rate": 1.3500000000000001e-05, "loss": 0.009, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9230769230769231, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9364161849710982, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06592454016208649, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.940127388535032, "eval_overall_precision": 0.9389312977099237, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2935, "eval_samples_per_second": 579.206, "eval_steps_per_second": 10.221, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.07905176281929016, "learning_rate": 1.3000000000000001e-05, "loss": 0.0097, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9257142857142857, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9418604651162791, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06314301490783691, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9413265306122449, "eval_overall_precision": 0.9413265306122449, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.295, "eval_samples_per_second": 576.232, "eval_steps_per_second": 10.169, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.3864988088607788, "learning_rate": 1.25e-05, "loss": 0.0096, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.9152542372881357, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8901098901098901, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9178470254957507, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9257142857142857, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06267941743135452, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.9363867684478372, "eval_overall_precision": 0.934010152284264, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.2925, "eval_samples_per_second": 581.225, "eval_steps_per_second": 10.257, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.1783769130706787, "learning_rate": 1.2e-05, "loss": 0.0085, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9204545454545454, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9310344827586207, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.0647546648979187, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9389312977099236, "eval_overall_precision": 0.9365482233502538, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2954, "eval_samples_per_second": 575.462, "eval_steps_per_second": 10.155, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.05055546015501022, "learning_rate": 1.1500000000000002e-05, "loss": 0.0081, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9199999999999999, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.936046511627907, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06592388451099396, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.9387755102040817, "eval_overall_precision": 0.9387755102040817, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.2953, "eval_samples_per_second": 575.722, "eval_steps_per_second": 10.16, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.554315984249115, "learning_rate": 1.1000000000000001e-05, "loss": 0.0089, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.930635838150289, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9583333333333334, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06508428603410721, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9435897435897437, "eval_overall_precision": 0.9484536082474226, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.2951, "eval_samples_per_second": 576.156, "eval_steps_per_second": 10.167, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.5241426825523376, "learning_rate": 1.05e-05, "loss": 0.0075, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9204545454545454, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9310344827586207, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06432995200157166, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9389312977099236, "eval_overall_precision": 0.9365482233502538, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2963, "eval_samples_per_second": 573.823, "eval_steps_per_second": 10.126, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.7520523071289062, "learning_rate": 1e-05, "loss": 0.009, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9265536723163842, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9010989010989011, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9322033898305084, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9375, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06452789157629013, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9453621346886911, "eval_overall_precision": 0.9417721518987342, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2956, "eval_samples_per_second": 575.01, "eval_steps_per_second": 10.147, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.17528299987316132, "learning_rate": 9.5e-06, "loss": 0.0103, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9199999999999999, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.936046511627907, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06373251974582672, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.937579617834395, "eval_overall_precision": 0.9363867684478372, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.2941, "eval_samples_per_second": 577.956, "eval_steps_per_second": 10.199, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.528049111366272, "learning_rate": 9e-06, "loss": 0.0085, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9283667621776504, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9473684210526315, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06377752125263214, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9413265306122449, "eval_overall_precision": 0.9413265306122449, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2975, "eval_samples_per_second": 571.366, "eval_steps_per_second": 10.083, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.6417689323425293, "learning_rate": 8.500000000000002e-06, "loss": 0.0081, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9257142857142857, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9418604651162791, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06370905786752701, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.940127388535032, "eval_overall_precision": 0.9389312977099237, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2941, "eval_samples_per_second": 578.062, "eval_steps_per_second": 10.201, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.5219071507453918, "learning_rate": 8.000000000000001e-06, "loss": 0.0073, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9222222222222223, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8829787234042553, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06307025998830795, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2949, "eval_samples_per_second": 576.398, "eval_steps_per_second": 10.172, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.5325189232826233, "learning_rate": 7.5e-06, "loss": 0.0067, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.9385474860335195, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9032258064516129, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9291784702549575, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9371428571428572, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06220288947224617, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9441624365482234, "eval_overall_precision": 0.9393939393939394, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.297, "eval_samples_per_second": 572.382, "eval_steps_per_second": 10.101, "step": 8160 }, { "epoch": 86.0, "grad_norm": 0.02902698516845703, "learning_rate": 7.000000000000001e-06, "loss": 0.0072, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9222222222222223, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8829787234042553, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9230769230769231, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9364161849710982, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06882063299417496, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9402795425667091, "eval_overall_precision": 0.9367088607594937, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2956, "eval_samples_per_second": 575.107, "eval_steps_per_second": 10.149, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.15740731358528137, "learning_rate": 6.5000000000000004e-06, "loss": 0.0071, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.9273743016759777, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8924731182795699, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9204545454545454, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9310344827586207, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06579603254795074, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9377382465057178, "eval_overall_precision": 0.9341772151898734, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2986, "eval_samples_per_second": 569.242, "eval_steps_per_second": 10.045, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.22595557570457458, "learning_rate": 6e-06, "loss": 0.0083, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9183098591549295, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9209039548022598, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06871578842401505, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.935361216730038, "eval_overall_precision": 0.929471032745592, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2991, "eval_samples_per_second": 568.332, "eval_steps_per_second": 10.029, "step": 8448 }, { "epoch": 89.0, "grad_norm": 1.0093448162078857, "learning_rate": 5.500000000000001e-06, "loss": 0.008, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9333333333333332, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8936170212765957, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9310344827586207, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9529411764705882, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06777916848659515, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9438775510204082, "eval_overall_precision": 0.9438775510204082, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.3032, "eval_samples_per_second": 560.764, "eval_steps_per_second": 9.896, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.08126522600650787, "learning_rate": 5e-06, "loss": 0.0058, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9121813031161473, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.92, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06813067942857742, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.9314720812182741, "eval_overall_precision": 0.9267676767676768, "eval_overall_recall": 0.9362244897959183, "eval_runtime": 0.2936, "eval_samples_per_second": 579.036, "eval_steps_per_second": 10.218, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.019634226337075233, "learning_rate": 4.5e-06, "loss": 0.0076, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.9325842696629213, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9021739130434783, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9239436619718311, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9265536723163842, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06484521180391312, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9404309252217997, "eval_overall_precision": 0.9345088161209067, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2966, "eval_samples_per_second": 573.166, "eval_steps_per_second": 10.115, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.5347074270248413, "learning_rate": 4.000000000000001e-06, "loss": 0.0068, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9183098591549295, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9209039548022598, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06565537303686142, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.935361216730038, "eval_overall_precision": 0.929471032745592, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2952, "eval_samples_per_second": 575.915, "eval_steps_per_second": 10.163, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.07370373606681824, "learning_rate": 3.5000000000000004e-06, "loss": 0.0067, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9183098591549295, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9209039548022598, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.0658806711435318, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.935361216730038, "eval_overall_precision": 0.929471032745592, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2966, "eval_samples_per_second": 573.134, "eval_steps_per_second": 10.114, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.3025619387626648, "learning_rate": 3e-06, "loss": 0.0084, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.9333333333333332, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8936170212765957, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06517026573419571, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9416243654822335, "eval_overall_precision": 0.9368686868686869, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2948, "eval_samples_per_second": 576.719, "eval_steps_per_second": 10.177, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.1495310217142105, "learning_rate": 2.5e-06, "loss": 0.0061, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.9273743016759777, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8924731182795699, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.92090395480226, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9261363636363636, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06470558792352676, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9378960709759189, "eval_overall_precision": 0.9319899244332494, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.295, "eval_samples_per_second": 576.364, "eval_steps_per_second": 10.171, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.40779590606689453, "learning_rate": 2.0000000000000003e-06, "loss": 0.0063, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9157303370786517, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9157303370786517, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06416351348161697, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9341772151898734, "eval_overall_precision": 0.9271356783919598, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2942, "eval_samples_per_second": 577.85, "eval_steps_per_second": 10.197, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.008076786994934082, "learning_rate": 1.5e-06, "loss": 0.0067, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9273743016759777, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8924731182795699, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9291784702549575, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9371428571428572, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06449466198682785, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9416243654822335, "eval_overall_precision": 0.9368686868686869, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2974, "eval_samples_per_second": 571.562, "eval_steps_per_second": 10.086, "step": 9312 }, { "epoch": 98.0, "grad_norm": 0.2699550688266754, "learning_rate": 1.0000000000000002e-06, "loss": 0.0072, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.9222222222222223, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8829787234042553, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06595491617918015, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2935, "eval_samples_per_second": 579.188, "eval_steps_per_second": 10.221, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.05176588520407677, "learning_rate": 5.000000000000001e-07, "loss": 0.0069, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9222222222222223, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8829787234042553, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.0654757097363472, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2978, "eval_samples_per_second": 570.864, "eval_steps_per_second": 10.074, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.014442157931625843, "learning_rate": 0.0, "loss": 0.0068, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9222222222222223, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8829787234042553, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06546389311552048, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2953, "eval_samples_per_second": 575.755, "eval_steps_per_second": 10.16, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 4315798421360676.0, "train_loss": 0.03753832100580136, "train_runtime": 578.6389, "train_samples_per_second": 264.586, "train_steps_per_second": 16.591 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 4315798421360676.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }