{ "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": 1.2515305280685425, "learning_rate": 4.9500000000000004e-05, "loss": 0.8344, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.0, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.0, "eval_LOCATION_recall": 0.0, "eval_ORGANIZATION_f1": 0.011627906976744186, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.16666666666666666, "eval_ORGANIZATION_recall": 0.006024096385542169, "eval_PERSON_f1": 0.0, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.0, "eval_PERSON_recall": 0.0, "eval_loss": 0.5347160696983337, "eval_overall_accuracy": 0.8348422496570644, "eval_overall_f1": 0.004878048780487805, "eval_overall_precision": 0.1111111111111111, "eval_overall_recall": 0.0024937655860349127, "eval_runtime": 0.2508, "eval_samples_per_second": 677.861, "eval_steps_per_second": 11.962, "step": 96 }, { "epoch": 2.0, "grad_norm": 0.9823935031890869, "learning_rate": 4.9e-05, "loss": 0.4476, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.21126760563380284, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.30612244897959184, "eval_LOCATION_recall": 0.16129032258064516, "eval_ORGANIZATION_f1": 0.4082687338501292, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.3574660633484163, "eval_ORGANIZATION_recall": 0.4759036144578313, "eval_PERSON_f1": 0.40633245382585753, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.32489451476793246, "eval_PERSON_recall": 0.5422535211267606, "eval_loss": 0.3088568150997162, "eval_overall_accuracy": 0.9001371742112483, "eval_overall_f1": 0.37665198237885467, "eval_overall_precision": 0.33727810650887574, "eval_overall_recall": 0.42643391521197005, "eval_runtime": 0.2472, "eval_samples_per_second": 687.668, "eval_steps_per_second": 12.135, "step": 192 }, { "epoch": 3.0, "grad_norm": 0.8617516160011292, "learning_rate": 4.85e-05, "loss": 0.2971, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.4221105527638191, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.39622641509433965, "eval_LOCATION_recall": 0.45161290322580644, "eval_ORGANIZATION_f1": 0.6307277628032345, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.5707317073170731, "eval_ORGANIZATION_recall": 0.7048192771084337, "eval_PERSON_f1": 0.7430340557275541, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.6629834254143646, "eval_PERSON_recall": 0.8450704225352113, "eval_loss": 0.21093440055847168, "eval_overall_accuracy": 0.943758573388203, "eval_overall_f1": 0.6248600223964166, "eval_overall_precision": 0.5670731707317073, "eval_overall_recall": 0.6957605985037406, "eval_runtime": 0.2473, "eval_samples_per_second": 687.546, "eval_steps_per_second": 12.133, "step": 288 }, { "epoch": 4.0, "grad_norm": 1.127282738685608, "learning_rate": 4.8e-05, "loss": 0.21, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.6169154228855721, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.5740740740740741, "eval_LOCATION_recall": 0.6666666666666666, "eval_ORGANIZATION_f1": 0.7329545454545454, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.6935483870967742, "eval_ORGANIZATION_recall": 0.7771084337349398, "eval_PERSON_f1": 0.9054054054054054, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.8701298701298701, "eval_PERSON_recall": 0.9436619718309859, "eval_loss": 0.1358155459165573, "eval_overall_accuracy": 0.9626886145404664, "eval_overall_f1": 0.7656065959952887, "eval_overall_precision": 0.7254464285714286, "eval_overall_recall": 0.8104738154613467, "eval_runtime": 0.2506, "eval_samples_per_second": 678.373, "eval_steps_per_second": 11.971, "step": 384 }, { "epoch": 5.0, "grad_norm": 0.8156549334526062, "learning_rate": 4.75e-05, "loss": 0.1444, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.7623762376237624, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7064220183486238, "eval_LOCATION_recall": 0.8279569892473119, "eval_ORGANIZATION_f1": 0.7567567567567567, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.7544910179640718, "eval_ORGANIZATION_recall": 0.7590361445783133, "eval_PERSON_f1": 0.9241379310344827, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9054054054054054, "eval_PERSON_recall": 0.9436619718309859, "eval_loss": 0.09743102639913559, "eval_overall_accuracy": 0.9700960219478738, "eval_overall_f1": 0.8169696969696969, "eval_overall_precision": 0.7948113207547169, "eval_overall_recall": 0.8403990024937655, "eval_runtime": 0.2481, "eval_samples_per_second": 685.336, "eval_steps_per_second": 12.094, "step": 480 }, { "epoch": 6.0, "grad_norm": 1.2419793605804443, "learning_rate": 4.7e-05, "loss": 0.1206, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.821256038647343, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7456140350877193, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.8176470588235294, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.7988505747126436, "eval_ORGANIZATION_recall": 0.8373493975903614, "eval_PERSON_f1": 0.9547038327526133, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9448275862068966, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.08841585367918015, "eval_overall_accuracy": 0.9739368998628258, "eval_overall_f1": 0.8657074340527579, "eval_overall_precision": 0.8337182448036952, "eval_overall_recall": 0.9002493765586035, "eval_runtime": 0.2471, "eval_samples_per_second": 687.919, "eval_steps_per_second": 12.14, "step": 576 }, { "epoch": 7.0, "grad_norm": 0.851949155330658, "learning_rate": 4.6500000000000005e-05, "loss": 0.1044, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.8431372549019608, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7747747747747747, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.838150289017341, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8055555555555556, "eval_ORGANIZATION_recall": 0.8734939759036144, "eval_PERSON_f1": 0.9550173010380623, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9387755102040817, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.08039457350969315, "eval_overall_accuracy": 0.9761316872427983, "eval_overall_f1": 0.8796185935637664, "eval_overall_precision": 0.8424657534246576, "eval_overall_recall": 0.9201995012468828, "eval_runtime": 0.2473, "eval_samples_per_second": 687.5, "eval_steps_per_second": 12.132, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.4981534779071808, "learning_rate": 4.600000000000001e-05, "loss": 0.0937, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.8374384236453202, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7727272727272727, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.8596491228070176, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8352272727272727, "eval_ORGANIZATION_recall": 0.8855421686746988, "eval_PERSON_f1": 0.951388888888889, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9383561643835616, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.07107817381620407, "eval_overall_accuracy": 0.9786008230452675, "eval_overall_f1": 0.885954381752701, "eval_overall_precision": 0.8541666666666666, "eval_overall_recall": 0.9201995012468828, "eval_runtime": 0.2472, "eval_samples_per_second": 687.784, "eval_steps_per_second": 12.137, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.8645637631416321, "learning_rate": 4.55e-05, "loss": 0.0906, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.883248730964467, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8365384615384616, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8538011695906433, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8295454545454546, "eval_ORGANIZATION_recall": 0.8795180722891566, "eval_PERSON_f1": 0.9550173010380623, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9387755102040817, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06350412964820862, "eval_overall_accuracy": 0.9805212620027435, "eval_overall_f1": 0.8961352657004832, "eval_overall_precision": 0.8688524590163934, "eval_overall_recall": 0.9251870324189526, "eval_runtime": 0.2464, "eval_samples_per_second": 690.046, "eval_steps_per_second": 12.177, "step": 864 }, { "epoch": 10.0, "grad_norm": 3.138153553009033, "learning_rate": 4.5e-05, "loss": 0.086, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.8686868686868686, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.819047619047619, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.8587896253602305, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8232044198895028, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.9448275862068966, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9256756756756757, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.06287253648042679, "eval_overall_accuracy": 0.980246913580247, "eval_overall_f1": 0.8910179640718562, "eval_overall_precision": 0.8571428571428571, "eval_overall_recall": 0.9276807980049875, "eval_runtime": 0.2499, "eval_samples_per_second": 680.176, "eval_steps_per_second": 12.003, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.7408162951469421, "learning_rate": 4.4500000000000004e-05, "loss": 0.0759, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.9157894736842105, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8969072164948454, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8612716763005781, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8277777777777777, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.9480968858131488, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9319727891156463, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.05623672530055046, "eval_overall_accuracy": 0.9807956104252401, "eval_overall_f1": 0.9042424242424242, "eval_overall_precision": 0.8797169811320755, "eval_overall_recall": 0.9301745635910225, "eval_runtime": 0.2493, "eval_samples_per_second": 681.943, "eval_steps_per_second": 12.034, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.7820143699645996, "learning_rate": 4.4000000000000006e-05, "loss": 0.071, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.8900523560209423, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8673469387755102, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.8767908309455588, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8360655737704918, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.951388888888889, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9383561643835616, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.053700800985097885, "eval_overall_accuracy": 0.9818930041152263, "eval_overall_f1": 0.9057971014492753, "eval_overall_precision": 0.8782201405152225, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2484, "eval_samples_per_second": 684.374, "eval_steps_per_second": 12.077, "step": 1152 }, { "epoch": 13.0, "grad_norm": 0.6278756260871887, "learning_rate": 4.35e-05, "loss": 0.0669, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.911917098445596, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.88, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8869047619047619, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8764705882352941, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.951388888888889, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9383561643835616, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.05038106441497803, "eval_overall_accuracy": 0.9840877914951989, "eval_overall_f1": 0.9155446756425949, "eval_overall_precision": 0.8990384615384616, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2483, "eval_samples_per_second": 684.535, "eval_steps_per_second": 12.08, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.803413987159729, "learning_rate": 4.3e-05, "loss": 0.0639, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.8934010152284263, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8461538461538461, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8862275449101796, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8809523809523809, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.9616724738675958, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9517241379310345, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05162254720926285, "eval_overall_accuracy": 0.9832647462277092, "eval_overall_f1": 0.9144254278728606, "eval_overall_precision": 0.8968824940047961, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2484, "eval_samples_per_second": 684.479, "eval_steps_per_second": 12.079, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.6189452409744263, "learning_rate": 4.25e-05, "loss": 0.061, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.90625, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8882352941176471, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.867816091954023, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.046318329870700836, "eval_overall_accuracy": 0.9854595336076818, "eval_overall_f1": 0.9193154034229828, "eval_overall_precision": 0.9016786570743405, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.248, "eval_samples_per_second": 685.602, "eval_steps_per_second": 12.099, "step": 1440 }, { "epoch": 16.0, "grad_norm": 6.981437683105469, "learning_rate": 4.2e-05, "loss": 0.0594, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.9025641025641027, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8627450980392157, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8882352941176471, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.867816091954023, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.047691043466329575, "eval_overall_accuracy": 0.9851851851851852, "eval_overall_f1": 0.9195121951219511, "eval_overall_precision": 0.8997613365155132, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2481, "eval_samples_per_second": 685.084, "eval_steps_per_second": 12.09, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.6753572821617126, "learning_rate": 4.15e-05, "loss": 0.0567, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.9025641025641027, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8627450980392157, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8797653958944281, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8571428571428571, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.046721987426280975, "eval_overall_accuracy": 0.9843621399176955, "eval_overall_f1": 0.9159561510353227, "eval_overall_precision": 0.8952380952380953, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2484, "eval_samples_per_second": 684.391, "eval_steps_per_second": 12.077, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.679158627986908, "learning_rate": 4.1e-05, "loss": 0.0526, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9263157894736842, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.896755162241888, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8786127167630058, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04115588217973709, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9310344827586207, "eval_overall_precision": 0.9197080291970803, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2481, "eval_samples_per_second": 685.229, "eval_steps_per_second": 12.092, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.7228365540504456, "learning_rate": 4.05e-05, "loss": 0.05, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9014925373134328, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.893491124260355, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04271138086915016, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9310344827586207, "eval_overall_precision": 0.9197080291970803, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2489, "eval_samples_per_second": 682.96, "eval_steps_per_second": 12.052, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.877856433391571, "learning_rate": 4e-05, "loss": 0.0474, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8725490196078431, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.8979591836734694, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8700564971751412, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04383207857608795, "eval_overall_accuracy": 0.9854595336076818, "eval_overall_f1": 0.9258809234507899, "eval_overall_precision": 0.9028436018957346, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2475, "eval_samples_per_second": 686.991, "eval_steps_per_second": 12.123, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.4687682092189789, "learning_rate": 3.9500000000000005e-05, "loss": 0.0472, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9026548672566371, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.884393063583815, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04151730239391327, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9326805385556914, "eval_overall_precision": 0.9158653846153846, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2474, "eval_samples_per_second": 687.161, "eval_steps_per_second": 12.126, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.6318132281303406, "learning_rate": 3.9000000000000006e-05, "loss": 0.0426, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.8914956011730205, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8685714285714285, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04162847250699997, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9255189255189255, "eval_overall_precision": 0.9066985645933014, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2478, "eval_samples_per_second": 686.167, "eval_steps_per_second": 12.109, "step": 2112 }, { "epoch": 23.0, "grad_norm": 0.8861734867095947, "learning_rate": 3.85e-05, "loss": 0.0422, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.9278350515463919, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9036144578313253, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9036144578313253, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04213300719857216, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9321824907521579, "eval_overall_precision": 0.9219512195121952, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2479, "eval_samples_per_second": 685.803, "eval_steps_per_second": 12.102, "step": 2208 }, { "epoch": 24.0, "grad_norm": 0.8380584120750427, "learning_rate": 3.8e-05, "loss": 0.0418, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8737864077669902, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9096385542168675, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9096385542168675, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04504883661866188, "eval_overall_accuracy": 0.9849108367626886, "eval_overall_f1": 0.932349323493235, "eval_overall_precision": 0.9199029126213593, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2469, "eval_samples_per_second": 688.558, "eval_steps_per_second": 12.151, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.7206587195396423, "learning_rate": 3.7500000000000003e-05, "loss": 0.038, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.896755162241888, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8786127167630058, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04219139367341995, "eval_overall_accuracy": 0.9851851851851852, "eval_overall_f1": 0.9266503667481663, "eval_overall_precision": 0.9088729016786571, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2474, "eval_samples_per_second": 687.15, "eval_steps_per_second": 12.126, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.24927875399589539, "learning_rate": 3.7e-05, "loss": 0.037, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.8914956011730205, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8685714285714285, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04009304568171501, "eval_overall_accuracy": 0.9857338820301783, "eval_overall_f1": 0.9243902439024391, "eval_overall_precision": 0.9045346062052506, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2476, "eval_samples_per_second": 686.552, "eval_steps_per_second": 12.116, "step": 2496 }, { "epoch": 27.0, "grad_norm": 0.8504116535186768, "learning_rate": 3.65e-05, "loss": 0.0346, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9014925373134328, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.893491124260355, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0395345501601696, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9287469287469288, "eval_overall_precision": 0.9152542372881356, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2466, "eval_samples_per_second": 689.337, "eval_steps_per_second": 12.165, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.6789959073066711, "learning_rate": 3.6e-05, "loss": 0.0363, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.883720930232558, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8539325842696629, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04273953288793564, "eval_overall_accuracy": 0.984636488340192, "eval_overall_f1": 0.9219512195121952, "eval_overall_precision": 0.9021479713603818, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2479, "eval_samples_per_second": 685.64, "eval_steps_per_second": 12.1, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.7642568349838257, "learning_rate": 3.55e-05, "loss": 0.0366, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9278350515463919, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9047619047619048, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04186627268791199, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9348093480934809, "eval_overall_precision": 0.9223300970873787, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2465, "eval_samples_per_second": 689.706, "eval_steps_per_second": 12.171, "step": 2784 }, { "epoch": 30.0, "grad_norm": 1.0052415132522583, "learning_rate": 3.5e-05, "loss": 0.0323, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9112426035502958, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8953488372093024, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.979020979020979, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9722222222222222, "eval_PERSON_recall": 0.9859154929577465, "eval_loss": 0.04114536568522453, "eval_overall_accuracy": 0.9857338820301783, "eval_overall_f1": 0.9375764993880048, "eval_overall_precision": 0.9206730769230769, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2476, "eval_samples_per_second": 686.476, "eval_steps_per_second": 12.114, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.183027446269989, "learning_rate": 3.45e-05, "loss": 0.0325, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9014925373134328, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.893491124260355, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03967192769050598, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9300613496932515, "eval_overall_precision": 0.9154589371980676, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2471, "eval_samples_per_second": 688.035, "eval_steps_per_second": 12.142, "step": 2976 }, { "epoch": 32.0, "grad_norm": 1.4418869018554688, "learning_rate": 3.4000000000000007e-05, "loss": 0.0295, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.8934911242603549, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.877906976744186, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04143485426902771, "eval_overall_accuracy": 0.9857338820301783, "eval_overall_f1": 0.926470588235294, "eval_overall_precision": 0.9108433734939759, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2479, "eval_samples_per_second": 685.81, "eval_steps_per_second": 12.103, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.7897229790687561, "learning_rate": 3.35e-05, "loss": 0.0288, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.9081632653061223, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8640776699029126, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.8948948948948949, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8922155688622755, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04245557263493538, "eval_overall_accuracy": 0.9843621399176955, "eval_overall_f1": 0.9238329238329237, "eval_overall_precision": 0.910411622276029, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2416, "eval_samples_per_second": 703.7, "eval_steps_per_second": 12.418, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.4726623594760895, "learning_rate": 3.3e-05, "loss": 0.0288, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8793103448275862, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.0404244102537632, "eval_overall_accuracy": 0.9857338820301783, "eval_overall_f1": 0.9326805385556914, "eval_overall_precision": 0.9158653846153846, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2464, "eval_samples_per_second": 689.921, "eval_steps_per_second": 12.175, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.3474993109703064, "learning_rate": 3.2500000000000004e-05, "loss": 0.0264, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.9319371727748691, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9047619047619048, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8941176470588236, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03973115608096123, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9334975369458128, "eval_overall_precision": 0.9221411192214112, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.248, "eval_samples_per_second": 685.383, "eval_steps_per_second": 12.095, "step": 3360 }, { "epoch": 36.0, "grad_norm": 1.1945281028747559, "learning_rate": 3.2000000000000005e-05, "loss": 0.028, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9085365853658537, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9197530864197531, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04313274100422859, "eval_overall_accuracy": 0.9860082304526749, "eval_overall_f1": 0.9331683168316831, "eval_overall_precision": 0.9262899262899262, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2479, "eval_samples_per_second": 685.786, "eval_steps_per_second": 12.102, "step": 3456 }, { "epoch": 37.0, "grad_norm": 0.7278014421463013, "learning_rate": 3.15e-05, "loss": 0.0245, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9053254437869822, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8895348837209303, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03931031376123428, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9312039312039312, "eval_overall_precision": 0.9176755447941889, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.243, "eval_samples_per_second": 699.702, "eval_steps_per_second": 12.348, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.2667478919029236, "learning_rate": 3.1e-05, "loss": 0.0238, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9161676646706587, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9107142857142857, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04241136088967323, "eval_overall_accuracy": 0.9865569272976681, "eval_overall_f1": 0.9384236453201971, "eval_overall_precision": 0.927007299270073, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2411, "eval_samples_per_second": 705.103, "eval_steps_per_second": 12.443, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.5108228921890259, "learning_rate": 3.05e-05, "loss": 0.0238, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04109891504049301, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9334975369458128, "eval_overall_precision": 0.9221411192214112, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2424, "eval_samples_per_second": 701.346, "eval_steps_per_second": 12.377, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.6643406748771667, "learning_rate": 3e-05, "loss": 0.0233, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9107142857142857, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04073956981301308, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9348093480934809, "eval_overall_precision": 0.9223300970873787, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2486, "eval_samples_per_second": 683.86, "eval_steps_per_second": 12.068, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.1256953775882721, "learning_rate": 2.95e-05, "loss": 0.023, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8941176470588235, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8735632183908046, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04030044376850128, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9289215686274509, "eval_overall_precision": 0.9132530120481928, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2413, "eval_samples_per_second": 704.479, "eval_steps_per_second": 12.432, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.6961055397987366, "learning_rate": 2.9e-05, "loss": 0.0217, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9036144578313253, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9036144578313253, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0416085347533226, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9308641975308642, "eval_overall_precision": 0.921760391198044, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2419, "eval_samples_per_second": 702.701, "eval_steps_per_second": 12.401, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.5346093773841858, "learning_rate": 2.8499999999999998e-05, "loss": 0.0213, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9085365853658537, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9197530864197531, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.042657654732465744, "eval_overall_accuracy": 0.9854595336076818, "eval_overall_f1": 0.9330024813895781, "eval_overall_precision": 0.928395061728395, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2418, "eval_samples_per_second": 703.017, "eval_steps_per_second": 12.406, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.3173430860042572, "learning_rate": 2.8000000000000003e-05, "loss": 0.0216, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9374999999999999, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.041986607015132904, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9382716049382714, "eval_overall_precision": 0.9290953545232273, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.246, "eval_samples_per_second": 690.948, "eval_steps_per_second": 12.193, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.7651618123054504, "learning_rate": 2.7500000000000004e-05, "loss": 0.02, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.9381443298969072, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.900990099009901, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9151515151515153, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9207317073170732, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.042184460908174515, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9394313967861558, "eval_overall_precision": 0.9313725490196079, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2464, "eval_samples_per_second": 689.881, "eval_steps_per_second": 12.174, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.2049635946750641, "learning_rate": 2.7000000000000002e-05, "loss": 0.0187, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.9381443298969072, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.900990099009901, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9009009009009009, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8982035928143712, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04194054752588272, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9346485819975339, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2483, "eval_samples_per_second": 684.692, "eval_steps_per_second": 12.083, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.023664146661758423, "learning_rate": 2.6500000000000004e-05, "loss": 0.0195, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.90625, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8994082840236686, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8837209302325582, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.041643284261226654, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9287469287469288, "eval_overall_precision": 0.9152542372881356, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.247, "eval_samples_per_second": 688.215, "eval_steps_per_second": 12.145, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.45325198769569397, "learning_rate": 2.6000000000000002e-05, "loss": 0.0183, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9278350515463919, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.042113061994314194, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9371146732429101, "eval_overall_precision": 0.926829268292683, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2495, "eval_samples_per_second": 681.244, "eval_steps_per_second": 12.022, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.5517528653144836, "learning_rate": 2.5500000000000003e-05, "loss": 0.0182, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9072164948453608, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8712871287128713, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9014925373134328, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.893491124260355, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04115647077560425, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.929889298892989, "eval_overall_precision": 0.9174757281553398, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2468, "eval_samples_per_second": 688.914, "eval_steps_per_second": 12.157, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.24822808802127838, "learning_rate": 2.5e-05, "loss": 0.0154, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.042449820786714554, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2458, "eval_samples_per_second": 691.489, "eval_steps_per_second": 12.203, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.6639029383659363, "learning_rate": 2.45e-05, "loss": 0.017, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8725490196078431, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.911854103343465, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9202453987730062, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.043647680431604385, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9356435643564357, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2464, "eval_samples_per_second": 690.069, "eval_steps_per_second": 12.178, "step": 4896 }, { "epoch": 52.0, "grad_norm": 1.0882995128631592, "learning_rate": 2.4e-05, "loss": 0.0165, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8737864077669902, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9161676646706587, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9107142857142857, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04369182139635086, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9385749385749386, "eval_overall_precision": 0.9249394673123487, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.247, "eval_samples_per_second": 688.268, "eval_steps_per_second": 12.146, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.49739882349967957, "learning_rate": 2.35e-05, "loss": 0.0161, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04599626362323761, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9394313967861558, "eval_overall_precision": 0.9313725490196079, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2463, "eval_samples_per_second": 690.223, "eval_steps_per_second": 12.18, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.7823913097381592, "learning_rate": 2.3000000000000003e-05, "loss": 0.0149, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9194029850746269, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9112426035502958, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.045275408774614334, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.942189421894219, "eval_overall_precision": 0.9296116504854369, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2408, "eval_samples_per_second": 706.087, "eval_steps_per_second": 12.46, "step": 5184 }, { "epoch": 55.0, "grad_norm": 1.4682822227478027, "learning_rate": 2.25e-05, "loss": 0.0149, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9174311926605505, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9316770186335404, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04617098718881607, "eval_overall_accuracy": 0.9860082304526749, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2416, "eval_samples_per_second": 703.654, "eval_steps_per_second": 12.417, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.5497621893882751, "learning_rate": 2.2000000000000003e-05, "loss": 0.0155, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8737864077669902, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.0445222333073616, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.937269372693727, "eval_overall_precision": 0.9247572815533981, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2473, "eval_samples_per_second": 687.334, "eval_steps_per_second": 12.129, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.4604448080062866, "learning_rate": 2.15e-05, "loss": 0.0151, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.9333333333333333, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8921568627450981, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9009009009009009, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8982035928143712, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04727301374077797, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9359605911330049, "eval_overall_precision": 0.9245742092457421, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2465, "eval_samples_per_second": 689.77, "eval_steps_per_second": 12.172, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.6147039532661438, "learning_rate": 2.1e-05, "loss": 0.0164, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04918331652879715, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9397293972939729, "eval_overall_precision": 0.9271844660194175, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2467, "eval_samples_per_second": 689.214, "eval_steps_per_second": 12.163, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.46650344133377075, "learning_rate": 2.05e-05, "loss": 0.0144, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9484536082474228, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9108910891089109, "eval_LOCATION_recall": 0.989247311827957, "eval_ORGANIZATION_f1": 0.9212121212121213, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.926829268292683, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04636125639081001, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9468479604449938, "eval_overall_precision": 0.9387254901960784, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2466, "eval_samples_per_second": 689.434, "eval_steps_per_second": 12.166, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.5204694271087646, "learning_rate": 2e-05, "loss": 0.014, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.9238578680203046, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.875, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9135802469135801, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9367088607594937, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04992961883544922, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9391304347826087, "eval_overall_precision": 0.9356435643564357, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2482, "eval_samples_per_second": 684.973, "eval_steps_per_second": 12.088, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.19278350472450256, "learning_rate": 1.9500000000000003e-05, "loss": 0.0137, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.9435897435897437, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9019607843137255, "eval_LOCATION_recall": 0.989247311827957, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.045230720192193985, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9432098765432098, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2464, "eval_samples_per_second": 689.801, "eval_steps_per_second": 12.173, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.7140735387802124, "learning_rate": 1.9e-05, "loss": 0.0132, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.9435897435897437, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9019607843137255, "eval_LOCATION_recall": 0.989247311827957, "eval_ORGANIZATION_f1": 0.9151515151515153, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9207317073170732, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04675097018480301, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9443757725587144, "eval_overall_precision": 0.9362745098039216, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2473, "eval_samples_per_second": 687.351, "eval_steps_per_second": 12.13, "step": 5952 }, { "epoch": 63.0, "grad_norm": 1.3302243947982788, "learning_rate": 1.85e-05, "loss": 0.0125, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9047619047619048, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.045600615441799164, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9385749385749386, "eval_overall_precision": 0.9249394673123487, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2466, "eval_samples_per_second": 689.391, "eval_steps_per_second": 12.166, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.34228718280792236, "learning_rate": 1.8e-05, "loss": 0.0131, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.9435897435897437, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9019607843137255, "eval_LOCATION_recall": 0.989247311827957, "eval_ORGANIZATION_f1": 0.9174311926605505, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9316770186335404, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04637756571173668, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9454094292803972, "eval_overall_precision": 0.9407407407407408, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2467, "eval_samples_per_second": 689.133, "eval_steps_per_second": 12.161, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.6925381422042847, "learning_rate": 1.75e-05, "loss": 0.0119, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.9435897435897437, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9019607843137255, "eval_LOCATION_recall": 0.989247311827957, "eval_ORGANIZATION_f1": 0.9146341463414634, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9259259259259259, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.046413522213697433, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9442379182156134, "eval_overall_precision": 0.9384236453201971, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2464, "eval_samples_per_second": 690.046, "eval_steps_per_second": 12.177, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.6568042039871216, "learning_rate": 1.7000000000000003e-05, "loss": 0.0135, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9333333333333333, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8921568627450981, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.911854103343465, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9202453987730062, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04535233974456787, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9405940594059405, "eval_overall_precision": 0.9336609336609336, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2465, "eval_samples_per_second": 689.641, "eval_steps_per_second": 12.17, "step": 6336 }, { "epoch": 67.0, "grad_norm": 4.428787708282471, "learning_rate": 1.65e-05, "loss": 0.0115, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9381443298969072, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.900990099009901, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04708201065659523, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9407407407407408, "eval_overall_precision": 0.9315403422982885, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2465, "eval_samples_per_second": 689.666, "eval_steps_per_second": 12.171, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.8564044237136841, "learning_rate": 1.6000000000000003e-05, "loss": 0.0124, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9333333333333333, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8921568627450981, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9151515151515153, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9207317073170732, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04680604860186577, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2474, "eval_samples_per_second": 687.214, "eval_steps_per_second": 12.127, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.3934555947780609, "learning_rate": 1.55e-05, "loss": 0.0124, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9161676646706587, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9107142857142857, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.046413201838731766, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.939877300613497, "eval_overall_precision": 0.9251207729468599, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2476, "eval_samples_per_second": 686.695, "eval_steps_per_second": 12.118, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.11730984598398209, "learning_rate": 1.5e-05, "loss": 0.0107, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.9387755102040817, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8932038834951457, "eval_LOCATION_recall": 0.989247311827957, "eval_ORGANIZATION_f1": 0.9240121580547112, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.046998534351587296, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9456790123456791, "eval_overall_precision": 0.9364303178484108, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2466, "eval_samples_per_second": 689.398, "eval_steps_per_second": 12.166, "step": 6720 }, { "epoch": 71.0, "grad_norm": 1.1789263486862183, "learning_rate": 1.45e-05, "loss": 0.011, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9333333333333333, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8921568627450981, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9281437125748503, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9226190476190477, "eval_ORGANIZATION_recall": 0.9337349397590361, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04763010889291763, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9459459459459459, "eval_overall_precision": 0.9322033898305084, "eval_overall_recall": 0.9600997506234414, "eval_runtime": 0.2467, "eval_samples_per_second": 689.203, "eval_steps_per_second": 12.162, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.37678804993629456, "learning_rate": 1.4000000000000001e-05, "loss": 0.0102, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.9387755102040817, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8932038834951457, "eval_LOCATION_recall": 0.989247311827957, "eval_ORGANIZATION_f1": 0.9179331306990882, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04846316576004028, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9443757725587144, "eval_overall_precision": 0.9362745098039216, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2471, "eval_samples_per_second": 688.114, "eval_steps_per_second": 12.143, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.5454061627388, "learning_rate": 1.3500000000000001e-05, "loss": 0.0096, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9387755102040817, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8932038834951457, "eval_LOCATION_recall": 0.989247311827957, "eval_ORGANIZATION_f1": 0.9268292682926829, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9382716049382716, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04851234704256058, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9480198019801981, "eval_overall_precision": 0.941031941031941, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2479, "eval_samples_per_second": 685.841, "eval_steps_per_second": 12.103, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.3533796966075897, "learning_rate": 1.3000000000000001e-05, "loss": 0.0107, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9235474006116209, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.937888198757764, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04719574749469757, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9442379182156134, "eval_overall_precision": 0.9384236453201971, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2464, "eval_samples_per_second": 690.028, "eval_steps_per_second": 12.177, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.44997307658195496, "learning_rate": 1.25e-05, "loss": 0.0119, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9174311926605505, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9316770186335404, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.050360169261693954, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9404466501240694, "eval_overall_precision": 0.9358024691358025, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2459, "eval_samples_per_second": 691.313, "eval_steps_per_second": 12.2, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.039321430027484894, "learning_rate": 1.2e-05, "loss": 0.0105, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9238578680203046, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.875, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9179331306990882, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.048801541328430176, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9407407407407408, "eval_overall_precision": 0.9315403422982885, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2463, "eval_samples_per_second": 690.152, "eval_steps_per_second": 12.179, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.0831478014588356, "learning_rate": 1.1500000000000002e-05, "loss": 0.0102, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9244712990936556, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9272727272727272, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05038721486926079, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9445129469790383, "eval_overall_precision": 0.9341463414634147, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2463, "eval_samples_per_second": 690.309, "eval_steps_per_second": 12.182, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.6893078684806824, "learning_rate": 1.1000000000000001e-05, "loss": 0.0098, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9435897435897437, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9019607843137255, "eval_LOCATION_recall": 0.989247311827957, "eval_ORGANIZATION_f1": 0.9207317073170733, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9320987654320988, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04842834919691086, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.946716232961586, "eval_overall_precision": 0.9408866995073891, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2471, "eval_samples_per_second": 687.861, "eval_steps_per_second": 12.139, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.12848801910877228, "learning_rate": 1.05e-05, "loss": 0.0089, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9151515151515153, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9207317073170732, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.046809710562229156, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9404466501240694, "eval_overall_precision": 0.9358024691358025, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2471, "eval_samples_per_second": 687.867, "eval_steps_per_second": 12.139, "step": 7584 }, { "epoch": 80.0, "grad_norm": 3.2775826454162598, "learning_rate": 1e-05, "loss": 0.0113, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9278350515463919, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.049325987696647644, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9382716049382714, "eval_overall_precision": 0.9290953545232273, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2475, "eval_samples_per_second": 686.733, "eval_steps_per_second": 12.119, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.5914605855941772, "learning_rate": 9.5e-06, "loss": 0.0093, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.9278350515463919, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9216867469879518, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9216867469879518, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04886525124311447, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9432098765432098, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2419, "eval_samples_per_second": 702.856, "eval_steps_per_second": 12.403, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.2993466854095459, "learning_rate": 9e-06, "loss": 0.0094, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9137055837563453, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8653846153846154, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9024390243902438, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9135802469135802, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.051948778331279755, "eval_overall_accuracy": 0.9865569272976681, "eval_overall_f1": 0.9320148331273177, "eval_overall_precision": 0.9240196078431373, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2463, "eval_samples_per_second": 690.317, "eval_steps_per_second": 12.182, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.09933167695999146, "learning_rate": 8.500000000000002e-06, "loss": 0.0104, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8737864077669902, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9235474006116209, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.937888198757764, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05083442106842995, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9417596034696407, "eval_overall_precision": 0.9359605911330049, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2462, "eval_samples_per_second": 690.502, "eval_steps_per_second": 12.185, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.6622636318206787, "learning_rate": 8.000000000000001e-06, "loss": 0.0098, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9238578680203046, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.875, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9174311926605505, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9316770186335404, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05110112950205803, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9405940594059405, "eval_overall_precision": 0.9336609336609336, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2465, "eval_samples_per_second": 689.751, "eval_steps_per_second": 12.172, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.9869118332862854, "learning_rate": 7.5e-06, "loss": 0.0092, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.9435897435897437, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9019607843137255, "eval_LOCATION_recall": 0.989247311827957, "eval_ORGANIZATION_f1": 0.9240121580547112, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04933422431349754, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9480198019801981, "eval_overall_precision": 0.941031941031941, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2471, "eval_samples_per_second": 688.063, "eval_steps_per_second": 12.142, "step": 8160 }, { "epoch": 86.0, "grad_norm": 0.034973397850990295, "learning_rate": 7.000000000000001e-06, "loss": 0.0099, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9268292682926829, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9382716049382716, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05023263022303581, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9442379182156134, "eval_overall_precision": 0.9384236453201971, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2467, "eval_samples_per_second": 689.159, "eval_steps_per_second": 12.162, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.7182714939117432, "learning_rate": 6.5000000000000004e-06, "loss": 0.0092, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8737864077669902, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9207317073170733, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9320987654320988, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04963819310069084, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9405940594059405, "eval_overall_precision": 0.9336609336609336, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2467, "eval_samples_per_second": 689.229, "eval_steps_per_second": 12.163, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.2559051513671875, "learning_rate": 6e-06, "loss": 0.0091, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9207317073170733, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9320987654320988, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04898100346326828, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9430693069306931, "eval_overall_precision": 0.9361179361179361, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2471, "eval_samples_per_second": 687.978, "eval_steps_per_second": 12.141, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.5433230996131897, "learning_rate": 5.500000000000001e-06, "loss": 0.0091, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9179331306990882, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.0497022420167923, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2465, "eval_samples_per_second": 689.666, "eval_steps_per_second": 12.171, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.11955895274877548, "learning_rate": 5e-06, "loss": 0.0093, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9435897435897437, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9019607843137255, "eval_LOCATION_recall": 0.989247311827957, "eval_ORGANIZATION_f1": 0.9179331306990882, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04867757856845856, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9455445544554455, "eval_overall_precision": 0.9385749385749386, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2468, "eval_samples_per_second": 688.94, "eval_steps_per_second": 12.158, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.285544216632843, "learning_rate": 4.5e-06, "loss": 0.0096, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.9333333333333333, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8921568627450981, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9207317073170733, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9320987654320988, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04998737946152687, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9442379182156134, "eval_overall_precision": 0.9384236453201971, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2467, "eval_samples_per_second": 689.04, "eval_steps_per_second": 12.16, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.2488878220319748, "learning_rate": 4.000000000000001e-06, "loss": 0.0088, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9146341463414634, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9259259259259259, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05062616989016533, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2475, "eval_samples_per_second": 686.784, "eval_steps_per_second": 12.12, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.6846613883972168, "learning_rate": 3.5000000000000004e-06, "loss": 0.0088, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9179331306990882, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05102866142988205, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2489, "eval_samples_per_second": 683.077, "eval_steps_per_second": 12.054, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.13117815554141998, "learning_rate": 3e-06, "loss": 0.0089, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.911854103343465, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9202453987730062, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05131078138947487, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9381188118811881, "eval_overall_precision": 0.9312039312039312, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2473, "eval_samples_per_second": 687.551, "eval_steps_per_second": 12.133, "step": 9024 }, { "epoch": 95.0, "grad_norm": 4.9423346519470215, "learning_rate": 2.5e-06, "loss": 0.0087, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9146341463414634, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9259259259259259, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.050910960882902145, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2474, "eval_samples_per_second": 687.188, "eval_steps_per_second": 12.127, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.1094447672367096, "learning_rate": 2.0000000000000003e-06, "loss": 0.0082, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.9278350515463919, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.911854103343465, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9202453987730062, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05064791440963745, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.247, "eval_samples_per_second": 688.205, "eval_steps_per_second": 12.145, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.014906022697687149, "learning_rate": 1.5e-06, "loss": 0.0093, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9212121212121213, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.926829268292683, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.051176413893699646, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9432098765432098, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2472, "eval_samples_per_second": 687.686, "eval_steps_per_second": 12.136, "step": 9312 }, { "epoch": 98.0, "grad_norm": 1.6294134855270386, "learning_rate": 1.0000000000000002e-06, "loss": 0.0083, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9212121212121213, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.926829268292683, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.050931550562381744, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9432098765432098, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2467, "eval_samples_per_second": 689.064, "eval_steps_per_second": 12.16, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.08063169568777084, "learning_rate": 5.000000000000001e-07, "loss": 0.0075, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9212121212121213, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.926829268292683, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05072704330086708, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9432098765432098, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.247, "eval_samples_per_second": 688.242, "eval_steps_per_second": 12.145, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.1378341019153595, "learning_rate": 0.0, "loss": 0.0094, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9212121212121213, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.926829268292683, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.050813283771276474, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9432098765432098, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2463, "eval_samples_per_second": 690.211, "eval_steps_per_second": 12.18, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 3811398039807054.0, "train_loss": 0.044552158347020546, "train_runtime": 560.5468, "train_samples_per_second": 273.126, "train_steps_per_second": 17.126 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 3811398039807054.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }