{ "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": 1.154296636581421, "learning_rate": 4.9500000000000004e-05, "loss": 0.8413, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.07339449541284404, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.6666666666666666, "eval_LOCATION_recall": 0.038834951456310676, "eval_ORGANIZATION_f1": 0.29314420803782504, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.24603174603174602, "eval_ORGANIZATION_recall": 0.36257309941520466, "eval_PERSON_f1": 0.3125, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.28662420382165604, "eval_PERSON_recall": 0.3435114503816794, "eval_loss": 0.38607850670814514, "eval_overall_accuracy": 0.8702016017674675, "eval_overall_f1": 0.2707317073170732, "eval_overall_precision": 0.2674698795180723, "eval_overall_recall": 0.2740740740740741, "eval_runtime": 0.2778, "eval_samples_per_second": 612.026, "eval_steps_per_second": 10.8, "step": 96 }, { "epoch": 2.0, "grad_norm": 0.9952401518821716, "learning_rate": 4.9e-05, "loss": 0.3574, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.42982456140350883, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.392, "eval_LOCATION_recall": 0.47572815533980584, "eval_ORGANIZATION_f1": 0.6571428571428573, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.6424581005586593, "eval_ORGANIZATION_recall": 0.672514619883041, "eval_PERSON_f1": 0.7054794520547945, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.639751552795031, "eval_PERSON_recall": 0.7862595419847328, "eval_loss": 0.2100333273410797, "eval_overall_accuracy": 0.9384148025407346, "eval_overall_f1": 0.6137931034482759, "eval_overall_precision": 0.5741935483870968, "eval_overall_recall": 0.6592592592592592, "eval_runtime": 0.2799, "eval_samples_per_second": 607.421, "eval_steps_per_second": 10.719, "step": 192 }, { "epoch": 3.0, "grad_norm": 0.8579609990119934, "learning_rate": 4.85e-05, "loss": 0.1857, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.7755102040816327, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8172043010752689, "eval_LOCATION_recall": 0.7378640776699029, "eval_ORGANIZATION_f1": 0.7899159663865545, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.7580645161290323, "eval_ORGANIZATION_recall": 0.8245614035087719, "eval_PERSON_f1": 0.9545454545454546, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9473684210526315, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.10320296883583069, "eval_overall_accuracy": 0.9674123170394918, "eval_overall_f1": 0.839657282741738, "eval_overall_precision": 0.8325242718446602, "eval_overall_recall": 0.8469135802469135, "eval_runtime": 0.2825, "eval_samples_per_second": 601.783, "eval_steps_per_second": 10.62, "step": 288 }, { "epoch": 4.0, "grad_norm": 1.0279494524002075, "learning_rate": 4.8e-05, "loss": 0.1252, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.8262910798122066, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8, "eval_LOCATION_recall": 0.8543689320388349, "eval_ORGANIZATION_f1": 0.8370786516853932, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8054054054054054, "eval_ORGANIZATION_recall": 0.8713450292397661, "eval_PERSON_f1": 0.9699248120300753, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9555555555555556, "eval_PERSON_recall": 0.9847328244274809, "eval_loss": 0.07209818065166473, "eval_overall_accuracy": 0.9748688207677437, "eval_overall_f1": 0.8766467065868263, "eval_overall_precision": 0.8511627906976744, "eval_overall_recall": 0.9037037037037037, "eval_runtime": 0.281, "eval_samples_per_second": 604.896, "eval_steps_per_second": 10.675, "step": 384 }, { "epoch": 5.0, "grad_norm": 0.528812825679779, "learning_rate": 4.75e-05, "loss": 0.1047, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.8768472906403941, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.8640776699029126, "eval_ORGANIZATION_f1": 0.8424068767908308, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8258426966292135, "eval_ORGANIZATION_recall": 0.8596491228070176, "eval_PERSON_f1": 0.9662921348314607, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9485294117647058, "eval_PERSON_recall": 0.9847328244274809, "eval_loss": 0.06120831146836281, "eval_overall_accuracy": 0.980115990057995, "eval_overall_f1": 0.8913308913308914, "eval_overall_precision": 0.8816425120772947, "eval_overall_recall": 0.9012345679012346, "eval_runtime": 0.2803, "eval_samples_per_second": 606.413, "eval_steps_per_second": 10.701, "step": 480 }, { "epoch": 6.0, "grad_norm": 0.8603781461715698, "learning_rate": 4.7e-05, "loss": 0.0891, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.817391304347826, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.7401574803149606, "eval_LOCATION_recall": 0.912621359223301, "eval_ORGANIZATION_f1": 0.836923076923077, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8831168831168831, "eval_ORGANIZATION_recall": 0.7953216374269005, "eval_PERSON_f1": 0.9847328244274809, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9847328244274809, "eval_PERSON_recall": 0.9847328244274809, "eval_loss": 0.06224534660577774, "eval_overall_accuracy": 0.9795636564484949, "eval_overall_f1": 0.8788249694002447, "eval_overall_precision": 0.8713592233009708, "eval_overall_recall": 0.8864197530864197, "eval_runtime": 0.2793, "eval_samples_per_second": 608.591, "eval_steps_per_second": 10.74, "step": 576 }, { "epoch": 7.0, "grad_norm": 1.19534432888031, "learning_rate": 4.6500000000000005e-05, "loss": 0.0819, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.8571428571428571, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9032258064516129, "eval_LOCATION_recall": 0.8155339805825242, "eval_ORGANIZATION_f1": 0.8795518207282913, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8440860215053764, "eval_ORGANIZATION_recall": 0.9181286549707602, "eval_PERSON_f1": 0.9809885931558935, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9772727272727273, "eval_PERSON_recall": 0.9847328244274809, "eval_loss": 0.05232277885079384, "eval_overall_accuracy": 0.9848108257387462, "eval_overall_f1": 0.9068627450980392, "eval_overall_precision": 0.9002433090024331, "eval_overall_recall": 0.9135802469135802, "eval_runtime": 0.28, "eval_samples_per_second": 607.137, "eval_steps_per_second": 10.714, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.39013275504112244, "learning_rate": 4.600000000000001e-05, "loss": 0.077, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.8815165876777251, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8611111111111112, "eval_LOCATION_recall": 0.9029126213592233, "eval_ORGANIZATION_f1": 0.8850574712643677, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8700564971751412, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9808429118773947, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9846153846153847, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.046886101365089417, "eval_overall_accuracy": 0.9859154929577465, "eval_overall_f1": 0.9146341463414634, "eval_overall_precision": 0.9036144578313253, "eval_overall_recall": 0.9259259259259259, "eval_runtime": 0.2794, "eval_samples_per_second": 608.403, "eval_steps_per_second": 10.737, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.584625780582428, "learning_rate": 4.55e-05, "loss": 0.0726, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.9029126213592235, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9029126213592233, "eval_LOCATION_recall": 0.9029126213592233, "eval_ORGANIZATION_f1": 0.8920454545454546, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8674033149171271, "eval_ORGANIZATION_recall": 0.9181286549707602, "eval_PERSON_f1": 0.9847328244274809, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9847328244274809, "eval_PERSON_recall": 0.9847328244274809, "eval_loss": 0.04818890616297722, "eval_overall_accuracy": 0.9859154929577465, "eval_overall_f1": 0.9243902439024391, "eval_overall_precision": 0.9132530120481928, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2786, "eval_samples_per_second": 610.235, "eval_steps_per_second": 10.769, "step": 864 }, { "epoch": 10.0, "grad_norm": 0.8925082683563232, "learning_rate": 4.5e-05, "loss": 0.0664, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.8609865470852018, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8, "eval_LOCATION_recall": 0.9320388349514563, "eval_ORGANIZATION_f1": 0.8495575221238938, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8571428571428571, "eval_ORGANIZATION_recall": 0.8421052631578947, "eval_PERSON_f1": 0.9847328244274809, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9847328244274809, "eval_PERSON_recall": 0.9847328244274809, "eval_loss": 0.05027275159955025, "eval_overall_accuracy": 0.9814968240817453, "eval_overall_f1": 0.8956310679611651, "eval_overall_precision": 0.8806682577565632, "eval_overall_recall": 0.9111111111111111, "eval_runtime": 0.2823, "eval_samples_per_second": 602.245, "eval_steps_per_second": 10.628, "step": 960 }, { "epoch": 11.0, "grad_norm": 1.0277563333511353, "learning_rate": 4.4500000000000004e-05, "loss": 0.0571, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.8666666666666667, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8504672897196262, "eval_LOCATION_recall": 0.883495145631068, "eval_ORGANIZATION_f1": 0.8786127167630058, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8685714285714285, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9808429118773947, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9846153846153847, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04224003106355667, "eval_overall_accuracy": 0.9861916597624966, "eval_overall_f1": 0.908200734394125, "eval_overall_precision": 0.9004854368932039, "eval_overall_recall": 0.9160493827160494, "eval_runtime": 0.28, "eval_samples_per_second": 607.053, "eval_steps_per_second": 10.713, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.6607075929641724, "learning_rate": 4.4000000000000006e-05, "loss": 0.0572, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.883495145631068, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.883495145631068, "eval_ORGANIZATION_f1": 0.8888888888888888, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8888888888888888, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.045749567449092865, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9135802469135802, "eval_overall_precision": 0.9135802469135802, "eval_overall_recall": 0.9135802469135802, "eval_runtime": 0.2801, "eval_samples_per_second": 606.902, "eval_steps_per_second": 10.71, "step": 1152 }, { "epoch": 13.0, "grad_norm": 1.4749391078948975, "learning_rate": 4.35e-05, "loss": 0.0526, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.8778280542986426, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8220338983050848, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.8787878787878787, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9119496855345912, "eval_ORGANIZATION_recall": 0.847953216374269, "eval_PERSON_f1": 0.9808429118773947, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9846153846153847, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.049907758831977844, "eval_overall_accuracy": 0.9831538249102458, "eval_overall_f1": 0.9113300492610836, "eval_overall_precision": 0.9090909090909091, "eval_overall_recall": 0.9135802469135802, "eval_runtime": 0.2872, "eval_samples_per_second": 591.879, "eval_steps_per_second": 10.445, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.45261910557746887, "learning_rate": 4.3e-05, "loss": 0.0525, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.9073170731707318, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9117647058823529, "eval_LOCATION_recall": 0.9029126213592233, "eval_ORGANIZATION_f1": 0.9032258064516129, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9808429118773947, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9846153846153847, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04711678624153137, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.929368029739777, "eval_overall_precision": 0.9328358208955224, "eval_overall_recall": 0.9259259259259259, "eval_runtime": 0.2808, "eval_samples_per_second": 605.38, "eval_steps_per_second": 10.683, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.8226550221443176, "learning_rate": 4.25e-05, "loss": 0.0496, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.9194312796208531, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8981481481481481, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.8934911242603552, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9041916167664671, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.9808429118773947, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9846153846153847, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.040322195738554, "eval_overall_accuracy": 0.9884009942004971, "eval_overall_f1": 0.928395061728395, "eval_overall_precision": 0.928395061728395, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2788, "eval_samples_per_second": 609.826, "eval_steps_per_second": 10.762, "step": 1440 }, { "epoch": 16.0, "grad_norm": 0.7743126749992371, "learning_rate": 4.2e-05, "loss": 0.046, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.9056603773584906, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8807339449541285, "eval_LOCATION_recall": 0.9320388349514563, "eval_ORGANIZATION_f1": 0.8621700879765396, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8647058823529412, "eval_ORGANIZATION_recall": 0.8596491228070176, "eval_PERSON_f1": 0.9808429118773947, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9846153846153847, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04395757243037224, "eval_overall_accuracy": 0.9856393261529964, "eval_overall_f1": 0.9115479115479117, "eval_overall_precision": 0.9070904645476773, "eval_overall_recall": 0.9160493827160494, "eval_runtime": 0.2803, "eval_samples_per_second": 606.526, "eval_steps_per_second": 10.703, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.3855823278427124, "learning_rate": 4.15e-05, "loss": 0.0451, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.9108910891089108, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9292929292929293, "eval_LOCATION_recall": 0.8932038834951457, "eval_ORGANIZATION_f1": 0.9116809116809117, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8888888888888888, "eval_ORGANIZATION_recall": 0.935672514619883, "eval_PERSON_f1": 0.9733840304182511, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9696969696969697, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04449866712093353, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9313725490196078, "eval_overall_precision": 0.9245742092457421, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2803, "eval_samples_per_second": 606.527, "eval_steps_per_second": 10.703, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.19643624126911163, "learning_rate": 4.1e-05, "loss": 0.0424, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.9082125603864736, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9038461538461539, "eval_LOCATION_recall": 0.912621359223301, "eval_ORGANIZATION_f1": 0.8965517241379309, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8813559322033898, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9808429118773947, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9846153846153847, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.039011258631944656, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.926470588235294, "eval_overall_precision": 0.9197080291970803, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2819, "eval_samples_per_second": 603.073, "eval_steps_per_second": 10.642, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.5639841556549072, "learning_rate": 4.05e-05, "loss": 0.0414, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.909952606635071, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.9320388349514563, "eval_ORGANIZATION_f1": 0.8982035928143711, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9202453987730062, "eval_ORGANIZATION_recall": 0.8771929824561403, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.042465586215257645, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.9244114002478315, "eval_overall_precision": 0.927860696517413, "eval_overall_recall": 0.9209876543209876, "eval_runtime": 0.2795, "eval_samples_per_second": 608.252, "eval_steps_per_second": 10.734, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.4266958236694336, "learning_rate": 4e-05, "loss": 0.0396, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.9178743961352658, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9134615384615384, "eval_LOCATION_recall": 0.9223300970873787, "eval_ORGANIZATION_f1": 0.9028571428571428, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.88268156424581, "eval_ORGANIZATION_recall": 0.9239766081871345, "eval_PERSON_f1": 0.9731800766283524, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9769230769230769, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04242739826440811, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9290953545232274, "eval_overall_precision": 0.9200968523002422, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2825, "eval_samples_per_second": 601.835, "eval_steps_per_second": 10.621, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.9739372730255127, "learning_rate": 3.9500000000000005e-05, "loss": 0.0374, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.8858447488584474, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8362068965517241, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.8708708708708708, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8950617283950617, "eval_ORGANIZATION_recall": 0.847953216374269, "eval_PERSON_f1": 0.9808429118773947, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9846153846153847, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.044551216065883636, "eval_overall_accuracy": 0.9850869925434963, "eval_overall_f1": 0.9102091020910209, "eval_overall_precision": 0.9068627450980392, "eval_overall_recall": 0.9135802469135802, "eval_runtime": 0.2802, "eval_samples_per_second": 606.815, "eval_steps_per_second": 10.708, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.5449288487434387, "learning_rate": 3.9000000000000006e-05, "loss": 0.0366, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.8981481481481481, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8584070796460177, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.8823529411764705, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8875739644970414, "eval_ORGANIZATION_recall": 0.8771929824561403, "eval_PERSON_f1": 0.9808429118773947, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9846153846153847, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.045186128467321396, "eval_overall_accuracy": 0.9859154929577465, "eval_overall_f1": 0.9179926560587516, "eval_overall_precision": 0.9101941747572816, "eval_overall_recall": 0.9259259259259259, "eval_runtime": 0.2808, "eval_samples_per_second": 605.383, "eval_steps_per_second": 10.683, "step": 2112 }, { "epoch": 23.0, "grad_norm": 0.7807333469390869, "learning_rate": 3.85e-05, "loss": 0.0347, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.9150943396226414, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8899082568807339, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.8988095238095237, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.9808429118773947, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9846153846153847, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.04340049624443054, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.9295426452410382, "eval_overall_precision": 0.9306930693069307, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2812, "eval_samples_per_second": 604.558, "eval_steps_per_second": 10.669, "step": 2208 }, { "epoch": 24.0, "grad_norm": 1.3717243671417236, "learning_rate": 3.8e-05, "loss": 0.0317, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9194312796208531, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8981481481481481, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.8994082840236686, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04294424131512642, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.9261083743842363, "eval_overall_precision": 0.9238329238329238, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2802, "eval_samples_per_second": 606.757, "eval_steps_per_second": 10.707, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.10902079939842224, "learning_rate": 3.7500000000000003e-05, "loss": 0.0312, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.8968609865470851, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8333333333333334, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.8624999999999999, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9261744966442953, "eval_ORGANIZATION_recall": 0.8070175438596491, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05561819672584534, "eval_overall_accuracy": 0.9842584921292461, "eval_overall_f1": 0.9068322981366459, "eval_overall_precision": 0.9125, "eval_overall_recall": 0.9012345679012346, "eval_runtime": 0.278, "eval_samples_per_second": 611.479, "eval_steps_per_second": 10.791, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.38614901900291443, "learning_rate": 3.7e-05, "loss": 0.0306, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9186602870813397, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9056603773584906, "eval_LOCATION_recall": 0.9320388349514563, "eval_ORGANIZATION_f1": 0.8914956011730205, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8941176470588236, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9808429118773947, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9846153846153847, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.03936714679002762, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9272503082614056, "eval_overall_precision": 0.9261083743842364, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2793, "eval_samples_per_second": 608.582, "eval_steps_per_second": 10.74, "step": 2496 }, { "epoch": 27.0, "grad_norm": 0.4033605754375458, "learning_rate": 3.65e-05, "loss": 0.0287, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.8990825688073394, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8521739130434782, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8606060606060606, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8930817610062893, "eval_ORGANIZATION_recall": 0.8304093567251462, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05052424222230911, "eval_overall_accuracy": 0.9845346589339962, "eval_overall_f1": 0.9061728395061729, "eval_overall_precision": 0.9061728395061729, "eval_overall_recall": 0.9061728395061729, "eval_runtime": 0.2803, "eval_samples_per_second": 606.546, "eval_steps_per_second": 10.704, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.3330176770687103, "learning_rate": 3.6e-05, "loss": 0.0278, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9353233830845772, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9591836734693877, "eval_LOCATION_recall": 0.912621359223301, "eval_ORGANIZATION_f1": 0.9221902017291066, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9090909090909091, "eval_ORGANIZATION_recall": 0.935672514619883, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.03438619524240494, "eval_overall_accuracy": 0.9906103286384976, "eval_overall_f1": 0.9407407407407408, "eval_overall_precision": 0.9407407407407408, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2832, "eval_samples_per_second": 600.27, "eval_steps_per_second": 10.593, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.6742222905158997, "learning_rate": 3.55e-05, "loss": 0.0276, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9116279069767441, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.875, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8708708708708708, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8950617283950617, "eval_ORGANIZATION_recall": 0.847953216374269, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.042776476591825485, "eval_overall_accuracy": 0.9848108257387462, "eval_overall_f1": 0.9135802469135802, "eval_overall_precision": 0.9135802469135802, "eval_overall_recall": 0.9135802469135802, "eval_runtime": 0.2831, "eval_samples_per_second": 600.503, "eval_steps_per_second": 10.597, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.2162758857011795, "learning_rate": 3.5e-05, "loss": 0.0261, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.9359605911330049, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.95, "eval_LOCATION_recall": 0.9223300970873787, "eval_ORGANIZATION_f1": 0.8914285714285715, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8715083798882681, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04228820279240608, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9253365973072215, "eval_overall_precision": 0.9174757281553398, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2813, "eval_samples_per_second": 604.358, "eval_steps_per_second": 10.665, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.4813241958618164, "learning_rate": 3.45e-05, "loss": 0.0262, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.9074074074074074, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8672566371681416, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8848484848484848, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9182389937106918, "eval_ORGANIZATION_recall": 0.8538011695906432, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.045065514743328094, "eval_overall_accuracy": 0.9853631593482464, "eval_overall_f1": 0.9160493827160494, "eval_overall_precision": 0.9160493827160494, "eval_overall_recall": 0.9160493827160494, "eval_runtime": 0.279, "eval_samples_per_second": 609.351, "eval_steps_per_second": 10.753, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.23773515224456787, "learning_rate": 3.4000000000000007e-05, "loss": 0.026, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9065420560747663, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8738738738738738, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9063444108761328, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9375, "eval_ORGANIZATION_recall": 0.8771929824561403, "eval_PERSON_f1": 0.9618320610687023, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9618320610687023, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.039794400334358215, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9244114002478315, "eval_overall_precision": 0.927860696517413, "eval_overall_recall": 0.9209876543209876, "eval_runtime": 0.2781, "eval_samples_per_second": 611.237, "eval_steps_per_second": 10.787, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.36368289589881897, "learning_rate": 3.35e-05, "loss": 0.0265, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.9186602870813397, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9056603773584906, "eval_LOCATION_recall": 0.9320388349514563, "eval_ORGANIZATION_f1": 0.8973607038123168, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9618320610687023, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9618320610687023, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.04403886944055557, "eval_overall_accuracy": 0.9861916597624966, "eval_overall_f1": 0.9236453201970444, "eval_overall_precision": 0.9213759213759214, "eval_overall_recall": 0.9259259259259259, "eval_runtime": 0.2825, "eval_samples_per_second": 601.774, "eval_steps_per_second": 10.62, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.22275644540786743, "learning_rate": 3.3e-05, "loss": 0.0258, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.9107981220657277, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8818181818181818, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9037900874635568, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9011627906976745, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9618320610687023, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9618320610687023, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.042414624243974686, "eval_overall_accuracy": 0.9859154929577465, "eval_overall_f1": 0.9242053789731051, "eval_overall_precision": 0.9152542372881356, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2779, "eval_samples_per_second": 611.719, "eval_steps_per_second": 10.795, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.5709071755409241, "learning_rate": 3.2500000000000004e-05, "loss": 0.0257, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.9201877934272301, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8909090909090909, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8656716417910448, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8841463414634146, "eval_ORGANIZATION_recall": 0.847953216374269, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.047432590276002884, "eval_overall_accuracy": 0.983982325324496, "eval_overall_f1": 0.9124537607891491, "eval_overall_precision": 0.9113300492610837, "eval_overall_recall": 0.9135802469135802, "eval_runtime": 0.2799, "eval_samples_per_second": 607.347, "eval_steps_per_second": 10.718, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.7812547087669373, "learning_rate": 3.2000000000000005e-05, "loss": 0.0249, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.8909090909090909, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8376068376068376, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8632218844984803, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8987341772151899, "eval_ORGANIZATION_recall": 0.8304093567251462, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04805381968617439, "eval_overall_accuracy": 0.9845346589339962, "eval_overall_f1": 0.902829028290283, "eval_overall_precision": 0.8995098039215687, "eval_overall_recall": 0.9061728395061729, "eval_runtime": 0.2796, "eval_samples_per_second": 608.006, "eval_steps_per_second": 10.73, "step": 3456 }, { "epoch": 37.0, "grad_norm": 0.3765964210033417, "learning_rate": 3.15e-05, "loss": 0.0232, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.9514563106796117, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9514563106796117, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8953488372093024, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8901734104046243, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.03953908383846283, "eval_overall_accuracy": 0.9886771610052472, "eval_overall_f1": 0.9323493234932348, "eval_overall_precision": 0.928921568627451, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2818, "eval_samples_per_second": 603.348, "eval_steps_per_second": 10.647, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.1870630830526352, "learning_rate": 3.1e-05, "loss": 0.0221, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9289099526066351, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9074074074074074, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.94375, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.9618320610687023, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9618320610687023, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.05032067000865936, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9328358208955223, "eval_overall_precision": 0.9398496240601504, "eval_overall_recall": 0.9259259259259259, "eval_runtime": 0.2799, "eval_samples_per_second": 607.423, "eval_steps_per_second": 10.719, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.309161514043808, "learning_rate": 3.05e-05, "loss": 0.0215, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.9282296650717704, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9150943396226415, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.8934911242603552, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9041916167664671, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.9618320610687023, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9618320610687023, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.04577593132853508, "eval_overall_accuracy": 0.9853631593482464, "eval_overall_f1": 0.9245982694684796, "eval_overall_precision": 0.9257425742574258, "eval_overall_recall": 0.9234567901234568, "eval_runtime": 0.2823, "eval_samples_per_second": 602.268, "eval_steps_per_second": 10.628, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.36333176493644714, "learning_rate": 3e-05, "loss": 0.0206, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9074626865671643, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.926829268292683, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9618320610687023, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9618320610687023, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.0451781339943409, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9333333333333333, "eval_overall_precision": 0.9333333333333333, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2791, "eval_samples_per_second": 609.089, "eval_steps_per_second": 10.749, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.660587728023529, "learning_rate": 2.95e-05, "loss": 0.0208, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.9423076923076923, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9333333333333333, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9581749049429659, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9545454545454546, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.05270150303840637, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.9341614906832298, "eval_overall_precision": 0.94, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2814, "eval_samples_per_second": 604.116, "eval_steps_per_second": 10.661, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.05708932504057884, "learning_rate": 2.9e-05, "loss": 0.0189, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.9428571428571428, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9252336448598131, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04301833361387253, "eval_overall_accuracy": 0.9884009942004971, "eval_overall_f1": 0.9379652605459058, "eval_overall_precision": 0.942643391521197, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2806, "eval_samples_per_second": 605.813, "eval_steps_per_second": 10.691, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.13359729945659637, "learning_rate": 2.8499999999999998e-05, "loss": 0.0196, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9171597633136095, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9281437125748503, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9618320610687023, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9618320610687023, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.04314638674259186, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9371146732429099, "eval_overall_precision": 0.9359605911330049, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2794, "eval_samples_per_second": 608.504, "eval_steps_per_second": 10.738, "step": 4128 }, { "epoch": 44.0, "grad_norm": 1.188427448272705, "learning_rate": 2.8000000000000003e-05, "loss": 0.0197, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9150943396226414, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8899082568807339, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.8705882352941177, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8757396449704142, "eval_ORGANIZATION_recall": 0.8654970760233918, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05358882248401642, "eval_overall_accuracy": 0.9831538249102458, "eval_overall_f1": 0.9128834355828221, "eval_overall_precision": 0.9073170731707317, "eval_overall_recall": 0.9185185185185185, "eval_runtime": 0.2804, "eval_samples_per_second": 606.293, "eval_steps_per_second": 10.699, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.41778749227523804, "learning_rate": 2.7500000000000004e-05, "loss": 0.0199, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.91324200913242, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8620689655172413, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.8807339449541285, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9230769230769231, "eval_ORGANIZATION_recall": 0.8421052631578947, "eval_PERSON_f1": 0.9581749049429659, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9545454545454546, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.05874410644173622, "eval_overall_accuracy": 0.983982325324496, "eval_overall_f1": 0.9147095179233622, "eval_overall_precision": 0.9158415841584159, "eval_overall_recall": 0.9135802469135802, "eval_runtime": 0.2805, "eval_samples_per_second": 606.039, "eval_steps_per_second": 10.695, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.5851004123687744, "learning_rate": 2.7000000000000002e-05, "loss": 0.0193, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.9124423963133641, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.868421052631579, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.8855421686746987, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9130434782608695, "eval_ORGANIZATION_recall": 0.8596491228070176, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.053020477294921875, "eval_overall_accuracy": 0.983982325324496, "eval_overall_f1": 0.9175891758917589, "eval_overall_precision": 0.9142156862745098, "eval_overall_recall": 0.9209876543209876, "eval_runtime": 0.2843, "eval_samples_per_second": 598.001, "eval_steps_per_second": 10.553, "step": 4416 }, { "epoch": 47.0, "grad_norm": 1.028637170791626, "learning_rate": 2.6500000000000004e-05, "loss": 0.0173, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.9468599033816425, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9423076923076923, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9005847953216374, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9005847953216374, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04588351398706436, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9323493234932348, "eval_overall_precision": 0.928921568627451, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2792, "eval_samples_per_second": 608.928, "eval_steps_per_second": 10.746, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.4050830900669098, "learning_rate": 2.6000000000000002e-05, "loss": 0.0197, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9345794392523364, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.8787878787878787, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9119496855345912, "eval_ORGANIZATION_recall": 0.847953216374269, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.059302061796188354, "eval_overall_accuracy": 0.9848108257387462, "eval_overall_f1": 0.9207920792079208, "eval_overall_precision": 0.9230769230769231, "eval_overall_recall": 0.9185185185185185, "eval_runtime": 0.2831, "eval_samples_per_second": 600.464, "eval_steps_per_second": 10.596, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.27565789222717285, "learning_rate": 2.5500000000000003e-05, "loss": 0.0164, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.8895522388059701, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9085365853658537, "eval_ORGANIZATION_recall": 0.8713450292397661, "eval_PERSON_f1": 0.9581749049429659, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9545454545454546, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.049104269593954086, "eval_overall_accuracy": 0.9859154929577465, "eval_overall_f1": 0.9245982694684796, "eval_overall_precision": 0.9257425742574258, "eval_overall_recall": 0.9234567901234568, "eval_runtime": 0.2811, "eval_samples_per_second": 604.686, "eval_steps_per_second": 10.671, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.41164466738700867, "learning_rate": 2.5e-05, "loss": 0.0174, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.9478672985781991, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9259259259259259, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.8999999999999999, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9053254437869822, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.047566816210746765, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.9336609336609337, "eval_overall_precision": 0.9290953545232273, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2834, "eval_samples_per_second": 599.851, "eval_steps_per_second": 10.586, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.09784907102584839, "learning_rate": 2.45e-05, "loss": 0.0176, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.9351851851851852, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8938053097345132, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9113149847094802, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9551282051282052, "eval_ORGANIZATION_recall": 0.8713450292397661, "eval_PERSON_f1": 0.9581749049429659, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9545454545454546, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.06214034557342529, "eval_overall_accuracy": 0.9848108257387462, "eval_overall_f1": 0.9330024813895781, "eval_overall_precision": 0.9376558603491272, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2789, "eval_samples_per_second": 609.472, "eval_steps_per_second": 10.755, "step": 4896 }, { "epoch": 52.0, "grad_norm": 0.6346091032028198, "learning_rate": 2.4e-05, "loss": 0.0162, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9473684210526316, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9339622641509434, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9047619047619047, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9212121212121213, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9581749049429659, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9545454545454546, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.05192515254020691, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9331683168316831, "eval_overall_precision": 0.9354838709677419, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.2823, "eval_samples_per_second": 602.093, "eval_steps_per_second": 10.625, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.17365430295467377, "learning_rate": 2.35e-05, "loss": 0.0161, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9282296650717704, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9150943396226415, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9011627906976746, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8959537572254336, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04691587761044502, "eval_overall_accuracy": 0.9861916597624966, "eval_overall_f1": 0.9277845777233782, "eval_overall_precision": 0.9199029126213593, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2805, "eval_samples_per_second": 606.132, "eval_steps_per_second": 10.696, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.0691240131855011, "learning_rate": 2.3000000000000003e-05, "loss": 0.0149, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.930232558139535, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8928571428571429, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9037900874635568, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9011627906976745, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04544233903288841, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.930572472594397, "eval_overall_precision": 0.9182692307692307, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.2773, "eval_samples_per_second": 613.11, "eval_steps_per_second": 10.82, "step": 5184 }, { "epoch": 55.0, "grad_norm": 0.6814279556274414, "learning_rate": 2.25e-05, "loss": 0.0167, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.9478672985781991, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9259259259259259, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9134328358208956, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9329268292682927, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04625613987445831, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9394313967861556, "eval_overall_precision": 0.9405940594059405, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2772, "eval_samples_per_second": 613.224, "eval_steps_per_second": 10.822, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.6387161016464233, "learning_rate": 2.2000000000000003e-05, "loss": 0.0143, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9528301886792453, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.926605504587156, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9249249249249248, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9506172839506173, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04298994317650795, "eval_overall_accuracy": 0.9884009942004971, "eval_overall_f1": 0.9455445544554455, "eval_overall_precision": 0.9478908188585607, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.2792, "eval_samples_per_second": 608.779, "eval_steps_per_second": 10.743, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.5601798892021179, "learning_rate": 2.15e-05, "loss": 0.015, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.9473684210526316, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9339622641509434, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9014925373134329, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9207317073170732, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04586603119969368, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9354838709677419, "eval_overall_precision": 0.940149625935162, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.2812, "eval_samples_per_second": 604.505, "eval_steps_per_second": 10.668, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.05784112215042114, "learning_rate": 2.1e-05, "loss": 0.0145, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.930232558139535, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8928571428571429, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9074626865671643, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.926829268292683, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04559914767742157, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.9334975369458127, "eval_overall_precision": 0.9312039312039312, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2821, "eval_samples_per_second": 602.701, "eval_steps_per_second": 10.636, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.602465808391571, "learning_rate": 2.05e-05, "loss": 0.0146, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9245283018867926, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8990825688073395, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9032258064516129, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04274475947022438, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9300613496932515, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2795, "eval_samples_per_second": 608.331, "eval_steps_per_second": 10.735, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.27823302149772644, "learning_rate": 2e-05, "loss": 0.0134, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.9488372093023257, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9107142857142857, "eval_LOCATION_recall": 0.9902912621359223, "eval_ORGANIZATION_f1": 0.9203539823008849, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9285714285714286, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04713748022913933, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.943627450980392, "eval_overall_precision": 0.9367396593673966, "eval_overall_recall": 0.9506172839506173, "eval_runtime": 0.2763, "eval_samples_per_second": 615.294, "eval_steps_per_second": 10.858, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.3316862881183624, "learning_rate": 1.9500000000000003e-05, "loss": 0.0147, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.9320388349514563, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9320388349514563, "eval_LOCATION_recall": 0.9320388349514563, "eval_ORGANIZATION_f1": 0.9064327485380118, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9064327485380117, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04435887932777405, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9310344827586207, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2751, "eval_samples_per_second": 617.93, "eval_steps_per_second": 10.905, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.8274697661399841, "learning_rate": 1.9e-05, "loss": 0.0135, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9043478260869565, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.896551724137931, "eval_ORGANIZATION_recall": 0.9122807017543859, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04778444021940231, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9328449328449329, "eval_overall_precision": 0.9227053140096618, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.2758, "eval_samples_per_second": 616.424, "eval_steps_per_second": 10.878, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.5358163714408875, "learning_rate": 1.85e-05, "loss": 0.0134, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.9333333333333335, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9158878504672897, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8999999999999999, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9053254437869822, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9770992366412213, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9770992366412213, "eval_PERSON_recall": 0.9770992366412213, "eval_loss": 0.05117568001151085, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9334975369458127, "eval_overall_precision": 0.9312039312039312, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2766, "eval_samples_per_second": 614.654, "eval_steps_per_second": 10.847, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.7101930379867554, "learning_rate": 1.8e-05, "loss": 0.0126, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9047619047619047, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9212121212121213, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.053059667348861694, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9333333333333333, "eval_overall_precision": 0.9333333333333333, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2857, "eval_samples_per_second": 595.046, "eval_steps_per_second": 10.501, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.5656111836433411, "learning_rate": 1.75e-05, "loss": 0.0123, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.9371980676328503, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9326923076923077, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.9112426035502958, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9221556886227545, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04893874004483223, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.934487021013597, "eval_overall_precision": 0.9356435643564357, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2788, "eval_samples_per_second": 609.728, "eval_steps_per_second": 10.76, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.493802547454834, "learning_rate": 1.7000000000000003e-05, "loss": 0.0136, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9483568075117371, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9181818181818182, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9166666666666666, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9333333333333333, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.050606537610292435, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.939729397293973, "eval_overall_precision": 0.9362745098039216, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.2802, "eval_samples_per_second": 606.636, "eval_steps_per_second": 10.705, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.12623465061187744, "learning_rate": 1.65e-05, "loss": 0.0118, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9483568075117371, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9181818181818182, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9134328358208956, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9329268292682927, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.049193013459444046, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.938423645320197, "eval_overall_precision": 0.9361179361179361, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.282, "eval_samples_per_second": 602.857, "eval_steps_per_second": 10.639, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.15083251893520355, "learning_rate": 1.6000000000000003e-05, "loss": 0.0114, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9488372093023257, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9107142857142857, "eval_LOCATION_recall": 0.9902912621359223, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9382716049382716, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05525859445333481, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.938423645320197, "eval_overall_precision": 0.9361179361179361, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2774, "eval_samples_per_second": 612.733, "eval_steps_per_second": 10.813, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.22303181886672974, "learning_rate": 1.55e-05, "loss": 0.0118, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.9483568075117371, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9181818181818182, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.918918918918919, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9444444444444444, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05852819234132767, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9407407407407408, "eval_overall_precision": 0.9407407407407408, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2785, "eval_samples_per_second": 610.326, "eval_steps_per_second": 10.77, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.11185706406831741, "learning_rate": 1.5e-05, "loss": 0.0114, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.9295774647887324, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.8908554572271385, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8988095238095238, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05141285061836243, "eval_overall_accuracy": 0.9859154929577465, "eval_overall_f1": 0.9240196078431373, "eval_overall_precision": 0.9172749391727494, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.2779, "eval_samples_per_second": 611.764, "eval_steps_per_second": 10.796, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.08216023445129395, "learning_rate": 1.45e-05, "loss": 0.0112, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.8947368421052632, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8947368421052632, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.051121409982442856, "eval_overall_accuracy": 0.9861916597624966, "eval_overall_f1": 0.9266503667481663, "eval_overall_precision": 0.9176755447941889, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2773, "eval_samples_per_second": 613.148, "eval_steps_per_second": 10.82, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.09036260098218918, "learning_rate": 1.4000000000000001e-05, "loss": 0.0096, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.9351851851851852, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8938053097345132, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.8996960486322189, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9367088607594937, "eval_ORGANIZATION_recall": 0.8654970760233918, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.06673559546470642, "eval_overall_accuracy": 0.9856393261529964, "eval_overall_f1": 0.9306930693069307, "eval_overall_precision": 0.9330024813895782, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2772, "eval_samples_per_second": 613.366, "eval_steps_per_second": 10.824, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.16665448248386383, "learning_rate": 1.3500000000000001e-05, "loss": 0.0113, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9158878504672898, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8828828828828829, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.9134328358208956, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9329268292682927, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9694656488549618, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9694656488549618, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.053004443645477295, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9321824907521578, "eval_overall_precision": 0.9310344827586207, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.276, "eval_samples_per_second": 616.01, "eval_steps_per_second": 10.871, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.01753900572657585, "learning_rate": 1.3000000000000001e-05, "loss": 0.0124, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9085545722713864, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9166666666666666, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05075749754905701, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9336609336609337, "eval_overall_precision": 0.9290953545232273, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2777, "eval_samples_per_second": 612.27, "eval_steps_per_second": 10.805, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.35307741165161133, "learning_rate": 1.25e-05, "loss": 0.0116, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9134328358208956, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9329268292682927, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.04960966110229492, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9358024691358025, "eval_overall_precision": 0.9358024691358025, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2765, "eval_samples_per_second": 614.917, "eval_steps_per_second": 10.851, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.9773866534233093, "learning_rate": 1.2e-05, "loss": 0.0116, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9011627906976746, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8959537572254336, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05280172824859619, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9315403422982885, "eval_overall_precision": 0.9225181598062954, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2775, "eval_samples_per_second": 612.72, "eval_steps_per_second": 10.813, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.08705956488847733, "learning_rate": 1.1500000000000002e-05, "loss": 0.0103, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.9357798165137615, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8869565217391304, "eval_LOCATION_recall": 0.9902912621359223, "eval_ORGANIZATION_f1": 0.9051987767584098, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9487179487179487, "eval_ORGANIZATION_recall": 0.8654970760233918, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.06570810079574585, "eval_overall_accuracy": 0.9864678265672466, "eval_overall_f1": 0.9331683168316831, "eval_overall_precision": 0.9354838709677419, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.2775, "eval_samples_per_second": 612.546, "eval_steps_per_second": 10.81, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.4064823389053345, "learning_rate": 1.1000000000000001e-05, "loss": 0.0106, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9150943396226414, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8899082568807339, "eval_LOCATION_recall": 0.941747572815534, "eval_ORGANIZATION_f1": 0.8967551622418879, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9047619047619048, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.053998857736587524, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9238329238329238, "eval_overall_precision": 0.9193154034229829, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2769, "eval_samples_per_second": 614.026, "eval_steps_per_second": 10.836, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.092685766518116, "learning_rate": 1.05e-05, "loss": 0.0089, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9483568075117371, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9181818181818182, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.918918918918919, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9444444444444444, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9657794676806084, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9621212121212122, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05562268942594528, "eval_overall_accuracy": 0.9884009942004971, "eval_overall_f1": 0.9419035846724351, "eval_overall_precision": 0.943069306930693, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2762, "eval_samples_per_second": 615.547, "eval_steps_per_second": 10.863, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.28838929533958435, "learning_rate": 1e-05, "loss": 0.0103, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9345794392523364, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9080118694362017, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9216867469879518, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.9618320610687023, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9618320610687023, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.05623379722237587, "eval_overall_accuracy": 0.9870201601767468, "eval_overall_f1": 0.9323493234932348, "eval_overall_precision": 0.928921568627451, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2778, "eval_samples_per_second": 612.02, "eval_steps_per_second": 10.8, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.26533594727516174, "learning_rate": 9.5e-06, "loss": 0.0089, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.9383886255924171, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.9037900874635568, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9011627906976745, "eval_ORGANIZATION_recall": 0.9064327485380117, "eval_PERSON_f1": 0.9618320610687023, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9618320610687023, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.052786171436309814, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9313725490196078, "eval_overall_precision": 0.9245742092457421, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2776, "eval_samples_per_second": 612.495, "eval_steps_per_second": 10.809, "step": 7776 }, { "epoch": 82.0, "grad_norm": 1.136555790901184, "learning_rate": 9e-06, "loss": 0.0096, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9439252336448598, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9099099099099099, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9014925373134329, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9207317073170732, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.9581749049429659, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9545454545454546, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.059047553688287735, "eval_overall_accuracy": 0.9859154929577465, "eval_overall_f1": 0.9310344827586207, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.9333333333333333, "eval_runtime": 0.2799, "eval_samples_per_second": 607.313, "eval_steps_per_second": 10.717, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.5659166574478149, "learning_rate": 8.500000000000002e-06, "loss": 0.0092, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9345794392523364, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.8955223880597015, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9146341463414634, "eval_ORGANIZATION_recall": 0.8771929824561403, "eval_PERSON_f1": 0.9618320610687023, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9618320610687023, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.05770380422472954, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9272503082614056, "eval_overall_precision": 0.9261083743842364, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2786, "eval_samples_per_second": 610.236, "eval_steps_per_second": 10.769, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.058192767202854156, "learning_rate": 8.000000000000001e-06, "loss": 0.0092, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9245283018867926, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.8990825688073395, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8862973760932944, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8837209302325582, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9581749049429659, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9545454545454546, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.056178800761699677, "eval_overall_accuracy": 0.9853631593482464, "eval_overall_f1": 0.9193154034229828, "eval_overall_precision": 0.910411622276029, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2772, "eval_samples_per_second": 613.383, "eval_steps_per_second": 10.824, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.755519688129425, "learning_rate": 7.5e-06, "loss": 0.0095, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.9333333333333335, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9158878504672897, "eval_LOCATION_recall": 0.9514563106796117, "eval_ORGANIZATION_f1": 0.8941176470588235, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.8994082840236687, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.9618320610687023, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9618320610687023, "eval_PERSON_recall": 0.9618320610687023, "eval_loss": 0.05738423392176628, "eval_overall_accuracy": 0.9861916597624966, "eval_overall_f1": 0.9261083743842363, "eval_overall_precision": 0.9238329238329238, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2787, "eval_samples_per_second": 609.938, "eval_steps_per_second": 10.764, "step": 8160 }, { "epoch": 86.0, "grad_norm": 0.06316640228033066, "learning_rate": 7.000000000000001e-06, "loss": 0.0084, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9339622641509433, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.908256880733945, "eval_LOCATION_recall": 0.9611650485436893, "eval_ORGANIZATION_f1": 0.8982035928143711, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9202453987730062, "eval_ORGANIZATION_recall": 0.8771929824561403, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05678955093026161, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.928395061728395, "eval_overall_precision": 0.928395061728395, "eval_overall_recall": 0.928395061728395, "eval_runtime": 0.2797, "eval_samples_per_second": 607.709, "eval_steps_per_second": 10.724, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.02561737596988678, "learning_rate": 6.5000000000000004e-06, "loss": 0.008, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.9483568075117371, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9181818181818182, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05619414895772934, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9371146732429099, "eval_overall_precision": 0.9359605911330049, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2797, "eval_samples_per_second": 607.771, "eval_steps_per_second": 10.725, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.022458516061306, "learning_rate": 6e-06, "loss": 0.0081, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9389671361502346, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9080118694362017, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9216867469879518, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.054451458156108856, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9336609336609337, "eval_overall_precision": 0.9290953545232273, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2778, "eval_samples_per_second": 611.945, "eval_steps_per_second": 10.799, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.13002897799015045, "learning_rate": 5.500000000000001e-06, "loss": 0.0096, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9345794392523364, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.9032258064516129, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.053021419793367386, "eval_overall_accuracy": 0.9867439933719967, "eval_overall_f1": 0.9304029304029304, "eval_overall_precision": 0.9202898550724637, "eval_overall_recall": 0.9407407407407408, "eval_runtime": 0.2764, "eval_samples_per_second": 614.952, "eval_steps_per_second": 10.852, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.33970993757247925, "learning_rate": 5e-06, "loss": 0.0081, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9345794392523364, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.8955223880597015, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9146341463414634, "eval_ORGANIZATION_recall": 0.8771929824561403, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.056356798857450485, "eval_overall_accuracy": 0.9859154929577465, "eval_overall_f1": 0.927429274292743, "eval_overall_precision": 0.9240196078431373, "eval_overall_recall": 0.9308641975308642, "eval_runtime": 0.2775, "eval_samples_per_second": 612.584, "eval_steps_per_second": 10.81, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.22719332575798035, "learning_rate": 4.5e-06, "loss": 0.0073, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.9345794392523364, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.970873786407767, "eval_ORGANIZATION_f1": 0.8999999999999999, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9053254437869822, "eval_ORGANIZATION_recall": 0.8947368421052632, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.056220073252916336, "eval_overall_accuracy": 0.9872963269814968, "eval_overall_f1": 0.9290953545232274, "eval_overall_precision": 0.9200968523002422, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2772, "eval_samples_per_second": 613.36, "eval_steps_per_second": 10.824, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.02064596675336361, "learning_rate": 4.000000000000001e-06, "loss": 0.0078, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9439252336448598, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9099099099099099, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9096385542168676, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.937888198757764, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.0590764619410038, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9358024691358025, "eval_overall_precision": 0.9358024691358025, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2759, "eval_samples_per_second": 616.17, "eval_steps_per_second": 10.874, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.32317450642585754, "learning_rate": 3.5000000000000004e-06, "loss": 0.0082, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9439252336448598, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9099099099099099, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.94375, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.060159020125865936, "eval_overall_accuracy": 0.9875724937862469, "eval_overall_f1": 0.9369592088998763, "eval_overall_precision": 0.9381188118811881, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2776, "eval_samples_per_second": 612.373, "eval_steps_per_second": 10.807, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.10917965322732925, "learning_rate": 3e-06, "loss": 0.0062, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.9483568075117371, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9181818181818182, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.913946587537092, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.927710843373494, "eval_ORGANIZATION_recall": 0.9005847953216374, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05830157920718193, "eval_overall_accuracy": 0.9884009942004971, "eval_overall_f1": 0.9385749385749386, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9432098765432099, "eval_runtime": 0.2778, "eval_samples_per_second": 612.004, "eval_steps_per_second": 10.8, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.6216530799865723, "learning_rate": 2.5e-06, "loss": 0.0083, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.9439252336448598, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9099099099099099, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05892162397503853, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9359605911330049, "eval_overall_precision": 0.9336609336609336, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2753, "eval_samples_per_second": 617.404, "eval_steps_per_second": 10.895, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.04704982042312622, "learning_rate": 2.0000000000000003e-06, "loss": 0.0072, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.9439252336448598, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9099099099099099, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.904191616766467, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.058384183794260025, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9334975369458127, "eval_overall_precision": 0.9312039312039312, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.2766, "eval_samples_per_second": 614.673, "eval_steps_per_second": 10.847, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.013867940753698349, "learning_rate": 1.5e-06, "loss": 0.009, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9439252336448598, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9099099099099099, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.904191616766467, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.8830409356725146, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05837693065404892, "eval_overall_accuracy": 0.987848660590997, "eval_overall_f1": 0.9334975369458127, "eval_overall_precision": 0.9312039312039312, "eval_overall_recall": 0.9358024691358025, "eval_runtime": 0.278, "eval_samples_per_second": 611.458, "eval_steps_per_second": 10.79, "step": 9312 }, { "epoch": 98.0, "grad_norm": 0.8402766585350037, "learning_rate": 1.0000000000000002e-06, "loss": 0.0066, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.9439252336448598, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9099099099099099, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.05868177488446236, "eval_overall_accuracy": 0.9884009942004971, "eval_overall_f1": 0.9359605911330049, "eval_overall_precision": 0.9336609336609336, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2768, "eval_samples_per_second": 614.108, "eval_steps_per_second": 10.837, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.030063997954130173, "learning_rate": 5.000000000000001e-07, "loss": 0.007, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9483568075117371, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9181818181818182, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.058369703590869904, "eval_overall_accuracy": 0.988124827395747, "eval_overall_f1": 0.9371146732429099, "eval_overall_precision": 0.9359605911330049, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2791, "eval_samples_per_second": 609.205, "eval_steps_per_second": 10.751, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.10819172859191895, "learning_rate": 0.0, "loss": 0.009, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9483568075117371, "eval_LOCATION_number": 103, "eval_LOCATION_precision": 0.9181818181818182, "eval_LOCATION_recall": 0.9805825242718447, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 171, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.8888888888888888, "eval_PERSON_f1": 0.962121212121212, "eval_PERSON_number": 131, "eval_PERSON_precision": 0.9548872180451128, "eval_PERSON_recall": 0.9694656488549618, "eval_loss": 0.058193083852529526, "eval_overall_accuracy": 0.9884009942004971, "eval_overall_f1": 0.9371146732429099, "eval_overall_precision": 0.9359605911330049, "eval_overall_recall": 0.9382716049382716, "eval_runtime": 0.2788, "eval_samples_per_second": 609.746, "eval_steps_per_second": 10.76, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 4325426924793372.0, "train_loss": 0.03781753840545813, "train_runtime": 546.8594, "train_samples_per_second": 279.962, "train_steps_per_second": 17.555 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 4325426924793372.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }