{ "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": 1.128080129623413, "learning_rate": 4.9500000000000004e-05, "loss": 0.8274, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.06201550387596899, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.3076923076923077, "eval_LOCATION_recall": 0.034482758620689655, "eval_ORGANIZATION_f1": 0.27477477477477474, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.21328671328671328, "eval_ORGANIZATION_recall": 0.3860759493670886, "eval_PERSON_f1": 0.2889733840304183, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.2733812949640288, "eval_PERSON_recall": 0.3064516129032258, "eval_loss": 0.38963502645492554, "eval_overall_accuracy": 0.865220971726599, "eval_overall_f1": 0.24641148325358853, "eval_overall_precision": 0.23515981735159816, "eval_overall_recall": 0.25879396984924624, "eval_runtime": 0.2864, "eval_samples_per_second": 593.655, "eval_steps_per_second": 10.476, "step": 96 }, { "epoch": 2.0, "grad_norm": 0.996844470500946, "learning_rate": 4.9e-05, "loss": 0.3677, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.4146341463414634, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.3923076923076923, "eval_LOCATION_recall": 0.4396551724137931, "eval_ORGANIZATION_f1": 0.5350318471337578, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.5384615384615384, "eval_ORGANIZATION_recall": 0.5316455696202531, "eval_PERSON_f1": 0.6576271186440678, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.5672514619883041, "eval_PERSON_recall": 0.782258064516129, "eval_loss": 0.21659503877162933, "eval_overall_accuracy": 0.9319242382651661, "eval_overall_f1": 0.5426900584795321, "eval_overall_precision": 0.5076586433260394, "eval_overall_recall": 0.5829145728643216, "eval_runtime": 0.2834, "eval_samples_per_second": 599.792, "eval_steps_per_second": 10.585, "step": 192 }, { "epoch": 3.0, "grad_norm": 0.7997106909751892, "learning_rate": 4.85e-05, "loss": 0.1945, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.787037037037037, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.85, "eval_LOCATION_recall": 0.7327586206896551, "eval_ORGANIZATION_f1": 0.7840909090909092, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.711340206185567, "eval_ORGANIZATION_recall": 0.8734177215189873, "eval_PERSON_f1": 0.9606299212598426, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9384615384615385, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.0953625813126564, "eval_overall_accuracy": 0.9706286027998902, "eval_overall_f1": 0.8394160583941606, "eval_overall_precision": 0.8136792452830188, "eval_overall_recall": 0.8668341708542714, "eval_runtime": 0.2859, "eval_samples_per_second": 594.529, "eval_steps_per_second": 10.492, "step": 288 }, { "epoch": 4.0, "grad_norm": 1.3715152740478516, "learning_rate": 4.8e-05, "loss": 0.1293, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.8477366255144033, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8110236220472441, "eval_LOCATION_recall": 0.8879310344827587, "eval_ORGANIZATION_f1": 0.8396501457725948, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.7783783783783784, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.08246859163045883, "eval_overall_accuracy": 0.9747460883886906, "eval_overall_f1": 0.8827751196172249, "eval_overall_precision": 0.8424657534246576, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2876, "eval_samples_per_second": 591.022, "eval_steps_per_second": 10.43, "step": 384 }, { "epoch": 5.0, "grad_norm": 0.7510249614715576, "learning_rate": 4.75e-05, "loss": 0.1104, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.8782608695652173, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8859649122807017, "eval_LOCATION_recall": 0.8706896551724138, "eval_ORGANIZATION_f1": 0.9051987767584098, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.8757396449704142, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.051644302904605865, "eval_overall_accuracy": 0.9849025528410651, "eval_overall_f1": 0.9230769230769231, "eval_overall_precision": 0.9117647058823529, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.29, "eval_samples_per_second": 586.308, "eval_steps_per_second": 10.347, "step": 480 }, { "epoch": 6.0, "grad_norm": 1.7944042682647705, "learning_rate": 4.7e-05, "loss": 0.096, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.846774193548387, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.7954545454545454, "eval_LOCATION_recall": 0.9051724137931034, "eval_ORGANIZATION_f1": 0.896103896103896, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.92, "eval_ORGANIZATION_recall": 0.8734177215189873, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.048994820564985275, "eval_overall_accuracy": 0.9829810595662916, "eval_overall_f1": 0.9059405940594059, "eval_overall_precision": 0.8926829268292683, "eval_overall_recall": 0.9195979899497487, "eval_runtime": 0.2892, "eval_samples_per_second": 587.763, "eval_steps_per_second": 10.372, "step": 576 }, { "epoch": 7.0, "grad_norm": 1.0523563623428345, "learning_rate": 4.6500000000000005e-05, "loss": 0.0809, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.8597285067873304, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9047619047619048, "eval_LOCATION_recall": 0.8189655172413793, "eval_ORGANIZATION_f1": 0.9036144578313253, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.8620689655172413, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04419120401144028, "eval_overall_accuracy": 0.9843535547625584, "eval_overall_f1": 0.9165628891656289, "eval_overall_precision": 0.908641975308642, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2901, "eval_samples_per_second": 585.942, "eval_steps_per_second": 10.34, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.6875613927841187, "learning_rate": 4.600000000000001e-05, "loss": 0.0757, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.9090909090909091, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.873015873015873, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9211356466876972, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9182389937106918, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04188111424446106, "eval_overall_accuracy": 0.9868240461158386, "eval_overall_f1": 0.9369592088998764, "eval_overall_precision": 0.9221411192214112, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2902, "eval_samples_per_second": 585.739, "eval_steps_per_second": 10.337, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.3846898376941681, "learning_rate": 4.55e-05, "loss": 0.0694, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.9113924050632911, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8925619834710744, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9657320872274143, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.950920245398773, "eval_ORGANIZATION_recall": 0.9810126582278481, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03360537439584732, "eval_overall_accuracy": 0.9917650288223991, "eval_overall_f1": 0.9554455445544553, "eval_overall_precision": 0.9414634146341463, "eval_overall_recall": 0.9698492462311558, "eval_runtime": 0.2876, "eval_samples_per_second": 591.023, "eval_steps_per_second": 10.43, "step": 864 }, { "epoch": 10.0, "grad_norm": 0.4912615418434143, "learning_rate": 4.5e-05, "loss": 0.0663, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.9142857142857144, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8682170542635659, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9419354838709677, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9605263157894737, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04144100472331047, "eval_overall_accuracy": 0.987098545155092, "eval_overall_f1": 0.943069306930693, "eval_overall_precision": 0.9292682926829269, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2903, "eval_samples_per_second": 585.525, "eval_steps_per_second": 10.333, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.2428193986415863, "learning_rate": 4.4500000000000004e-05, "loss": 0.0625, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.9211618257261411, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.888, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9426751592356688, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9487179487179487, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0336020290851593, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9490683229813666, "eval_overall_precision": 0.9385749385749386, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2889, "eval_samples_per_second": 588.414, "eval_steps_per_second": 10.384, "step": 1056 }, { "epoch": 12.0, "grad_norm": 1.136364221572876, "learning_rate": 4.4000000000000006e-05, "loss": 0.0567, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.9004329004329004, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9043478260869565, "eval_LOCATION_recall": 0.896551724137931, "eval_ORGANIZATION_f1": 0.9493670886075949, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9493670886075949, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04105456918478012, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9435382685069008, "eval_overall_precision": 0.9423558897243107, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2901, "eval_samples_per_second": 585.968, "eval_steps_per_second": 10.341, "step": 1152 }, { "epoch": 13.0, "grad_norm": 0.6924769282341003, "learning_rate": 4.35e-05, "loss": 0.0546, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.9184549356223175, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9145299145299145, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.9426751592356688, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9487179487179487, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.032622117549180984, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9485570890840652, "eval_overall_precision": 0.9473684210526315, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2894, "eval_samples_per_second": 587.496, "eval_steps_per_second": 10.368, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.4599927067756653, "learning_rate": 4.3e-05, "loss": 0.0473, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.9243697478991596, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9016393442622951, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9415584415584415, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9666666666666667, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9721115537848605, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9606299212598425, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.038140587508678436, "eval_overall_accuracy": 0.9881965413121054, "eval_overall_f1": 0.9460476787954831, "eval_overall_precision": 0.9448621553884712, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2906, "eval_samples_per_second": 584.96, "eval_steps_per_second": 10.323, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.6044618487358093, "learning_rate": 4.25e-05, "loss": 0.0478, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.9282700421940928, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9446254071661238, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9731543624161074, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.030891025438904762, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9521410579345088, "eval_overall_precision": 0.9545454545454546, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2878, "eval_samples_per_second": 590.746, "eval_steps_per_second": 10.425, "step": 1440 }, { "epoch": 16.0, "grad_norm": 0.6948764324188232, "learning_rate": 4.2e-05, "loss": 0.0436, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.935064935064935, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9391304347826087, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9496855345911951, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.94375, "eval_ORGANIZATION_recall": 0.9556962025316456, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03264456242322922, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9561952440550688, "eval_overall_precision": 0.9526184538653366, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2985, "eval_samples_per_second": 569.497, "eval_steps_per_second": 10.05, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.9365695118904114, "learning_rate": 4.15e-05, "loss": 0.0426, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.9256198347107438, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.935064935064935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.96, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9644268774703558, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9457364341085271, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.0373193733394146, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9414694894146949, "eval_overall_precision": 0.9333333333333333, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2893, "eval_samples_per_second": 587.582, "eval_steps_per_second": 10.369, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.23583665490150452, "learning_rate": 4.1e-05, "loss": 0.0399, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.031957585364580154, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9523809523809523, "eval_overall_precision": 0.95, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2861, "eval_samples_per_second": 594.251, "eval_steps_per_second": 10.487, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.6568490862846375, "learning_rate": 4.05e-05, "loss": 0.0373, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.935064935064935, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9391304347826087, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9493670886075949, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9493670886075949, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0372890867292881, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9560853199498118, "eval_overall_precision": 0.9548872180451128, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2884, "eval_samples_per_second": 589.488, "eval_steps_per_second": 10.403, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.44850486516952515, "learning_rate": 4e-05, "loss": 0.0354, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.927038626609442, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9230769230769231, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.033323984593153, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.950943396226415, "eval_overall_precision": 0.9521410579345088, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2921, "eval_samples_per_second": 582.088, "eval_steps_per_second": 10.272, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.6164305210113525, "learning_rate": 3.9500000000000005e-05, "loss": 0.0372, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9419354838709677, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9605263157894737, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.033080458641052246, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9558638083228247, "eval_overall_precision": 0.959493670886076, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2897, "eval_samples_per_second": 586.848, "eval_steps_per_second": 10.356, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.49077439308166504, "learning_rate": 3.9000000000000006e-05, "loss": 0.034, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.9310344827586207, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9310344827586207, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9493670886075949, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9493670886075949, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03408174589276314, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9548872180451128, "eval_overall_precision": 0.9525, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2885, "eval_samples_per_second": 589.247, "eval_steps_per_second": 10.398, "step": 2112 }, { "epoch": 23.0, "grad_norm": 0.29559168219566345, "learning_rate": 3.85e-05, "loss": 0.0331, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9396825396825397, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9426751592356688, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.035869140177965164, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9523809523809523, "eval_overall_precision": 0.95, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2882, "eval_samples_per_second": 589.803, "eval_steps_per_second": 10.408, "step": 2208 }, { "epoch": 24.0, "grad_norm": 0.44689151644706726, "learning_rate": 3.8e-05, "loss": 0.032, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9396551724137931, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9396551724137931, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9523809523809524, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9554140127388535, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.036276694387197495, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.958594730238394, "eval_overall_precision": 0.9573934837092731, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2848, "eval_samples_per_second": 596.88, "eval_steps_per_second": 10.533, "step": 2304 }, { "epoch": 25.0, "grad_norm": 1.1862599849700928, "learning_rate": 3.7500000000000003e-05, "loss": 0.0298, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.9487179487179486, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.940677966101695, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9641693811074918, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9932885906040269, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.042767103761434555, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9622166246851386, "eval_overall_precision": 0.9646464646464646, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2863, "eval_samples_per_second": 593.862, "eval_steps_per_second": 10.48, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.05711984261870384, "learning_rate": 3.7e-05, "loss": 0.0311, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9415584415584415, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9666666666666667, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.042612478137016296, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9508196721311476, "eval_overall_precision": 0.9544303797468354, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2867, "eval_samples_per_second": 592.985, "eval_steps_per_second": 10.464, "step": 2496 }, { "epoch": 27.0, "grad_norm": 0.9682767391204834, "learning_rate": 3.65e-05, "loss": 0.0307, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9514563106796117, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9735099337748344, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.038912009447813034, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9546599496221662, "eval_overall_precision": 0.9570707070707071, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2891, "eval_samples_per_second": 588.071, "eval_steps_per_second": 10.378, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.5806172490119934, "learning_rate": 3.6e-05, "loss": 0.0274, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9152542372881356, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04392585903406143, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9484276729559749, "eval_overall_precision": 0.9496221662468514, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2859, "eval_samples_per_second": 594.619, "eval_steps_per_second": 10.493, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.33236050605773926, "learning_rate": 3.55e-05, "loss": 0.0254, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9456066945606694, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9186991869918699, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04087366163730621, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9550000000000001, "eval_overall_precision": 0.9502487562189055, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2864, "eval_samples_per_second": 593.632, "eval_steps_per_second": 10.476, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.8669499158859253, "learning_rate": 3.5e-05, "loss": 0.0277, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.9344978165938865, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9469026548672567, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.9532710280373832, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9386503067484663, "eval_ORGANIZATION_recall": 0.9683544303797469, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03460657224059105, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9551122194513715, "eval_overall_precision": 0.948019801980198, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2848, "eval_samples_per_second": 596.901, "eval_steps_per_second": 10.534, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.7216441631317139, "learning_rate": 3.45e-05, "loss": 0.0267, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03778526932001114, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2846, "eval_samples_per_second": 597.335, "eval_steps_per_second": 10.541, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.6263610124588013, "learning_rate": 3.4000000000000007e-05, "loss": 0.026, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9573770491803278, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9931972789115646, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04354654252529144, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.962121212121212, "eval_overall_precision": 0.9670050761421319, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2853, "eval_samples_per_second": 595.796, "eval_steps_per_second": 10.514, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.7626103162765503, "learning_rate": 3.35e-05, "loss": 0.0241, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9511400651465798, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9798657718120806, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.036125801503658295, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.959493670886076, "eval_overall_precision": 0.9668367346938775, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2915, "eval_samples_per_second": 583.128, "eval_steps_per_second": 10.29, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.3419962525367737, "learning_rate": 3.3e-05, "loss": 0.0245, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.9523809523809523, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9565217391304348, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03828663378953934, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9596977329974811, "eval_overall_precision": 0.9621212121212122, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.285, "eval_samples_per_second": 596.582, "eval_steps_per_second": 10.528, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.5194388628005981, "learning_rate": 3.2500000000000004e-05, "loss": 0.0228, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.9523809523809523, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9565217391304348, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9548387096774194, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9736842105263158, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9721115537848605, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9606299212598425, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04244256764650345, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9595959595959597, "eval_overall_precision": 0.9644670050761421, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.287, "eval_samples_per_second": 592.427, "eval_steps_per_second": 10.455, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.5117161273956299, "learning_rate": 3.2000000000000005e-05, "loss": 0.0218, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.9491525423728815, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9333333333333333, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9415584415584415, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9666666666666667, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9644268774703558, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9457364341085271, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04172789677977562, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9510664993726474, "eval_overall_precision": 0.949874686716792, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2901, "eval_samples_per_second": 586.061, "eval_steps_per_second": 10.342, "step": 3456 }, { "epoch": 37.0, "grad_norm": 0.8829843401908875, "learning_rate": 3.15e-05, "loss": 0.0221, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.9487179487179486, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.940677966101695, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9721115537848605, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9606299212598425, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04155246913433075, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9547738693467337, "eval_overall_precision": 0.9547738693467337, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2869, "eval_samples_per_second": 592.631, "eval_steps_per_second": 10.458, "step": 3552 }, { "epoch": 38.0, "grad_norm": 1.269371747970581, "learning_rate": 3.1e-05, "loss": 0.0233, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9411764705882353, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.972972972972973, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04456571862101555, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9558638083228247, "eval_overall_precision": 0.959493670886076, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2844, "eval_samples_per_second": 597.765, "eval_steps_per_second": 10.549, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.20994959771633148, "learning_rate": 3.05e-05, "loss": 0.0197, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9329073482428115, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9419354838709677, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.041857898235321045, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9486858573216519, "eval_overall_precision": 0.9451371571072319, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2846, "eval_samples_per_second": 597.4, "eval_steps_per_second": 10.542, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.20928826928138733, "learning_rate": 3e-05, "loss": 0.0203, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.9437229437229439, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9478260869565217, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9396825396825397, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9426751592356688, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03918270766735077, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.951188986232791, "eval_overall_precision": 0.9476309226932669, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2837, "eval_samples_per_second": 599.134, "eval_steps_per_second": 10.573, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.0683557316660881, "learning_rate": 2.95e-05, "loss": 0.0191, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.9396551724137931, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9396551724137931, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9589905362776026, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9559748427672956, "eval_ORGANIZATION_recall": 0.9620253164556962, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.033662233501672745, "eval_overall_accuracy": 0.9923140269009059, "eval_overall_f1": 0.958801498127341, "eval_overall_precision": 0.9528535980148883, "eval_overall_recall": 0.964824120603015, "eval_runtime": 0.2856, "eval_samples_per_second": 595.211, "eval_steps_per_second": 10.504, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.38020259141921997, "learning_rate": 2.9e-05, "loss": 0.017, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9260450160771704, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9411764705882353, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9644268774703558, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9457364341085271, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03714501112699509, "eval_overall_accuracy": 0.9879220422728521, "eval_overall_f1": 0.9398496240601504, "eval_overall_precision": 0.9375, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2848, "eval_samples_per_second": 596.968, "eval_steps_per_second": 10.535, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.2194911390542984, "learning_rate": 2.8499999999999998e-05, "loss": 0.0174, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9437229437229439, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9478260869565217, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03944186493754387, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9546599496221662, "eval_overall_precision": 0.9570707070707071, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2889, "eval_samples_per_second": 588.456, "eval_steps_per_second": 10.385, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.0995759516954422, "learning_rate": 2.8000000000000003e-05, "loss": 0.018, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9198312236286921, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9008264462809917, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.045028068125247955, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9461827284105131, "eval_overall_precision": 0.942643391521197, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2859, "eval_samples_per_second": 594.686, "eval_steps_per_second": 10.494, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.47930288314819336, "learning_rate": 2.7500000000000004e-05, "loss": 0.018, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9419354838709677, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9605263157894737, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.039240602403879166, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.949748743718593, "eval_overall_precision": 0.949748743718593, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2876, "eval_samples_per_second": 591.172, "eval_steps_per_second": 10.432, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.600913941860199, "learning_rate": 2.7000000000000002e-05, "loss": 0.0167, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.927038626609442, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9230769230769231, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9358974358974359, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.948051948051948, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04149045795202255, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9460476787954831, "eval_overall_precision": 0.9448621553884712, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2835, "eval_samples_per_second": 599.622, "eval_steps_per_second": 10.582, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.4069179892539978, "learning_rate": 2.6500000000000004e-05, "loss": 0.0167, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.9396551724137931, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9396551724137931, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9545454545454545, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.98, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.040939461439847946, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9570707070707072, "eval_overall_precision": 0.9619289340101523, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2859, "eval_samples_per_second": 594.576, "eval_steps_per_second": 10.493, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.34376901388168335, "learning_rate": 2.6000000000000002e-05, "loss": 0.0156, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9220779220779222, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9466666666666667, "eval_ORGANIZATION_recall": 0.8987341772151899, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04170544445514679, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.947103274559194, "eval_overall_precision": 0.9494949494949495, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2864, "eval_samples_per_second": 593.667, "eval_steps_per_second": 10.476, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.07029540836811066, "learning_rate": 2.5500000000000003e-05, "loss": 0.0167, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9487179487179488, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.961038961038961, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04137362167239189, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9598997493734336, "eval_overall_precision": 0.9575, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2855, "eval_samples_per_second": 595.434, "eval_steps_per_second": 10.508, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.553500235080719, "learning_rate": 2.5e-05, "loss": 0.0153, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.9446808510638298, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9327731092436975, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9548387096774194, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9736842105263158, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.040293995290994644, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.958594730238394, "eval_overall_precision": 0.9573934837092731, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2899, "eval_samples_per_second": 586.33, "eval_steps_per_second": 10.347, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.4216265082359314, "learning_rate": 2.45e-05, "loss": 0.0175, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03513796627521515, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2864, "eval_samples_per_second": 593.633, "eval_steps_per_second": 10.476, "step": 4896 }, { "epoch": 52.0, "grad_norm": 0.6645966172218323, "learning_rate": 2.4e-05, "loss": 0.015, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9551282051282052, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9675324675324676, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03831518441438675, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9563046192259675, "eval_overall_precision": 0.9503722084367245, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2864, "eval_samples_per_second": 593.549, "eval_steps_per_second": 10.474, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.39954090118408203, "learning_rate": 2.35e-05, "loss": 0.0156, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.948051948051948, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9733333333333334, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04088861867785454, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9510664993726474, "eval_overall_precision": 0.949874686716792, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2894, "eval_samples_per_second": 587.486, "eval_steps_per_second": 10.367, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.7162357568740845, "learning_rate": 2.3000000000000003e-05, "loss": 0.0142, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.925764192139738, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9380530973451328, "eval_LOCATION_recall": 0.9137931034482759, "eval_ORGANIZATION_f1": 0.95, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9382716049382716, "eval_ORGANIZATION_recall": 0.9620253164556962, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03942888602614403, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9525, "eval_overall_precision": 0.9477611940298507, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2863, "eval_samples_per_second": 593.692, "eval_steps_per_second": 10.477, "step": 5184 }, { "epoch": 55.0, "grad_norm": 0.6623092293739319, "learning_rate": 2.25e-05, "loss": 0.0165, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.9446808510638298, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9327731092436975, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03844790160655975, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9536921151439299, "eval_overall_precision": 0.9501246882793017, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2852, "eval_samples_per_second": 595.981, "eval_steps_per_second": 10.517, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.4061862528324127, "learning_rate": 2.2000000000000003e-05, "loss": 0.0138, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9551282051282052, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9675324675324676, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04114224389195442, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9573934837092732, "eval_overall_precision": 0.955, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2852, "eval_samples_per_second": 596.109, "eval_steps_per_second": 10.52, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.38007456064224243, "learning_rate": 2.15e-05, "loss": 0.0136, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.9276595744680851, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9159663865546218, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.042864855378866196, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9498746867167919, "eval_overall_precision": 0.9475, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2836, "eval_samples_per_second": 599.417, "eval_steps_per_second": 10.578, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.11870250850915909, "learning_rate": 2.1e-05, "loss": 0.0136, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04294950142502785, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9548872180451128, "eval_overall_precision": 0.9525, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2899, "eval_samples_per_second": 586.399, "eval_steps_per_second": 10.348, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.5874012112617493, "learning_rate": 2.05e-05, "loss": 0.0138, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9237288135593221, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9083333333333333, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.926517571884984, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9354838709677419, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03952578455209732, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9425, "eval_overall_precision": 0.9378109452736318, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2832, "eval_samples_per_second": 600.224, "eval_steps_per_second": 10.592, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.09649311751127243, "learning_rate": 2e-05, "loss": 0.013, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.9382716049382717, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8976377952755905, "eval_LOCATION_recall": 0.9827586206896551, "eval_ORGANIZATION_f1": 0.9225806451612903, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9407894736842105, "eval_ORGANIZATION_recall": 0.9050632911392406, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.047314103692770004, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.946450809464508, "eval_overall_precision": 0.9382716049382716, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2833, "eval_samples_per_second": 600.001, "eval_steps_per_second": 10.588, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.6758869290351868, "learning_rate": 1.9500000000000003e-05, "loss": 0.0128, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9367088607594937, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9367088607594937, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03929971903562546, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.946450809464508, "eval_overall_precision": 0.9382716049382716, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2835, "eval_samples_per_second": 599.679, "eval_steps_per_second": 10.583, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.3419683277606964, "learning_rate": 1.9e-05, "loss": 0.011, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03987028822302818, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9611041405269761, "eval_overall_precision": 0.9598997493734336, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2871, "eval_samples_per_second": 592.108, "eval_steps_per_second": 10.449, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.023923255503177643, "learning_rate": 1.85e-05, "loss": 0.0121, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.9367088607594937, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9173553719008265, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9644268774703558, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9457364341085271, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04410358890891075, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9476309226932669, "eval_overall_precision": 0.9405940594059405, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2836, "eval_samples_per_second": 599.371, "eval_steps_per_second": 10.577, "step": 6048 }, { "epoch": 64.0, "grad_norm": 1.8180116415023804, "learning_rate": 1.8e-05, "loss": 0.0111, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9426751592356688, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9487179487179487, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9644268774703558, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9457364341085271, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04472411051392555, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.95, "eval_overall_precision": 0.945273631840796, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2873, "eval_samples_per_second": 591.73, "eval_steps_per_second": 10.442, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.6038116812705994, "learning_rate": 1.75e-05, "loss": 0.0117, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9682539682539683, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.953125, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04310223087668419, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9486858573216519, "eval_overall_precision": 0.9451371571072319, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2853, "eval_samples_per_second": 595.779, "eval_steps_per_second": 10.514, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.22309881448745728, "learning_rate": 1.7000000000000003e-05, "loss": 0.011, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9456066945606694, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9186991869918699, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9514563106796117, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9735099337748344, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9644268774703558, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9457364341085271, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.049640677869319916, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9538077403245941, "eval_overall_precision": 0.9478908188585607, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2827, "eval_samples_per_second": 601.253, "eval_steps_per_second": 10.61, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.8597439527511597, "learning_rate": 1.65e-05, "loss": 0.0112, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9358974358974359, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.948051948051948, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9644268774703558, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9457364341085271, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.046545740216970444, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9450000000000001, "eval_overall_precision": 0.9402985074626866, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2849, "eval_samples_per_second": 596.746, "eval_steps_per_second": 10.531, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.7216902375221252, "learning_rate": 1.6000000000000003e-05, "loss": 0.011, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9685039370078741, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9461538461538461, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04642572999000549, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9488139825218478, "eval_overall_precision": 0.9429280397022333, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2856, "eval_samples_per_second": 595.268, "eval_steps_per_second": 10.505, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.6276118159294128, "learning_rate": 1.55e-05, "loss": 0.0095, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.9276595744680851, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9159663865546218, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.045306939631700516, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9488139825218478, "eval_overall_precision": 0.9429280397022333, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.287, "eval_samples_per_second": 592.305, "eval_steps_per_second": 10.452, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.3850950300693512, "learning_rate": 1.5e-05, "loss": 0.0118, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.9198312236286921, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9008264462809917, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9493670886075949, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9493670886075949, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.048267751932144165, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9478908188585609, "eval_overall_precision": 0.9362745098039216, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2882, "eval_samples_per_second": 589.88, "eval_steps_per_second": 10.41, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.11313530802726746, "learning_rate": 1.45e-05, "loss": 0.0111, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.046932876110076904, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9561952440550688, "eval_overall_precision": 0.9526184538653366, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2852, "eval_samples_per_second": 595.98, "eval_steps_per_second": 10.517, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.010440285317599773, "learning_rate": 1.4000000000000001e-05, "loss": 0.0104, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.9276595744680851, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9159663865546218, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9419354838709677, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9605263157894737, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.047372810542583466, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9473684210526316, "eval_overall_precision": 0.945, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2838, "eval_samples_per_second": 598.973, "eval_steps_per_second": 10.57, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.15582597255706787, "learning_rate": 1.3500000000000001e-05, "loss": 0.0104, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9415584415584415, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9666666666666667, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9723320158102766, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9534883720930233, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04453054443001747, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9485570890840652, "eval_overall_precision": 0.9473684210526315, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2849, "eval_samples_per_second": 596.702, "eval_steps_per_second": 10.53, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.2509544789791107, "learning_rate": 1.3000000000000001e-05, "loss": 0.0115, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9644268774703558, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9457364341085271, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04860513284802437, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9473684210526316, "eval_overall_precision": 0.945, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2863, "eval_samples_per_second": 593.7, "eval_steps_per_second": 10.477, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.320895791053772, "learning_rate": 1.25e-05, "loss": 0.0109, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.9446808510638298, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9327731092436975, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9682539682539683, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.953125, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.050710711628198624, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9523809523809523, "eval_overall_precision": 0.95, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.284, "eval_samples_per_second": 598.6, "eval_steps_per_second": 10.564, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.19377972185611725, "learning_rate": 1.2e-05, "loss": 0.0094, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9327731092436975, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9098360655737705, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9612903225806452, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9802631578947368, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.052319228649139404, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9575, "eval_overall_precision": 0.9527363184079602, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2829, "eval_samples_per_second": 600.931, "eval_steps_per_second": 10.605, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.2946006655693054, "learning_rate": 1.1500000000000002e-05, "loss": 0.0107, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.9367088607594937, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9173553719008265, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9290322580645162, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9473684210526315, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9721115537848605, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9606299212598425, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04820028319954872, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9448621553884713, "eval_overall_precision": 0.9425, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2873, "eval_samples_per_second": 591.651, "eval_steps_per_second": 10.441, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.39715683460235596, "learning_rate": 1.1000000000000001e-05, "loss": 0.0095, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9487179487179488, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.961038961038961, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9644268774703558, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9457364341085271, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.053150177001953125, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.951188986232791, "eval_overall_precision": 0.9476309226932669, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2839, "eval_samples_per_second": 598.889, "eval_steps_per_second": 10.569, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.029291601851582527, "learning_rate": 1.05e-05, "loss": 0.0097, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04591898247599602, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9548872180451128, "eval_overall_precision": 0.9525, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2825, "eval_samples_per_second": 601.874, "eval_steps_per_second": 10.621, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.2444496750831604, "learning_rate": 1e-05, "loss": 0.0089, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9152542372881356, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0503772608935833, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9473684210526316, "eval_overall_precision": 0.945, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2864, "eval_samples_per_second": 593.6, "eval_steps_per_second": 10.475, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.04975898563861847, "learning_rate": 9.5e-06, "loss": 0.0095, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9333333333333332, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9363057324840764, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04681336507201195, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9486858573216519, "eval_overall_precision": 0.9451371571072319, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2843, "eval_samples_per_second": 597.979, "eval_steps_per_second": 10.553, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.036560989916324615, "learning_rate": 9e-06, "loss": 0.0091, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9514563106796117, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9735099337748344, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04577646404504776, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2908, "eval_samples_per_second": 584.608, "eval_steps_per_second": 10.317, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.1825098991394043, "learning_rate": 8.500000000000002e-06, "loss": 0.0084, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9511400651465798, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9798657718120806, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05219333991408348, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9558638083228247, "eval_overall_precision": 0.959493670886076, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2834, "eval_samples_per_second": 599.862, "eval_steps_per_second": 10.586, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.05879203975200653, "learning_rate": 8.000000000000001e-06, "loss": 0.009, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9511400651465798, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9798657718120806, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05156485736370087, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9559748427672956, "eval_overall_precision": 0.9571788413098237, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2858, "eval_samples_per_second": 594.892, "eval_steps_per_second": 10.498, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.5078015327453613, "learning_rate": 7.5e-06, "loss": 0.0075, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9389067524115756, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.954248366013072, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04952356219291687, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9486858573216519, "eval_overall_precision": 0.9451371571072319, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.283, "eval_samples_per_second": 600.77, "eval_steps_per_second": 10.602, "step": 8160 }, { "epoch": 86.0, "grad_norm": 0.052184294909238815, "learning_rate": 7.000000000000001e-06, "loss": 0.0089, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9237288135593221, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9083333333333333, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04943705350160599, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.9485570890840652, "eval_overall_precision": 0.9473684210526315, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2844, "eval_samples_per_second": 597.844, "eval_steps_per_second": 10.55, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.2575075626373291, "learning_rate": 6.5000000000000004e-06, "loss": 0.0087, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9358974358974359, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.948051948051948, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9682539682539683, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.953125, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.05178801715373993, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9450000000000001, "eval_overall_precision": 0.9402985074626866, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2865, "eval_samples_per_second": 593.278, "eval_steps_per_second": 10.47, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.9075930714607239, "learning_rate": 6e-06, "loss": 0.0096, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9237288135593221, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9083333333333333, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.935064935064935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.96, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9682539682539683, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.953125, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04699193686246872, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.9422110552763819, "eval_overall_precision": 0.9422110552763819, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2846, "eval_samples_per_second": 597.419, "eval_steps_per_second": 10.543, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.38378816843032837, "learning_rate": 5.500000000000001e-06, "loss": 0.0081, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9358974358974359, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.948051948051948, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9682539682539683, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.953125, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04642374813556671, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9450000000000001, "eval_overall_precision": 0.9402985074626866, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2866, "eval_samples_per_second": 593.247, "eval_steps_per_second": 10.469, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.013585579581558704, "learning_rate": 5e-06, "loss": 0.0083, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9419354838709677, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9605263157894737, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9682539682539683, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.953125, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04675604775547981, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9473684210526316, "eval_overall_precision": 0.945, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2914, "eval_samples_per_second": 583.338, "eval_steps_per_second": 10.294, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.21433831751346588, "learning_rate": 4.5e-06, "loss": 0.0079, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9152542372881356, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9396825396825397, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9426751592356688, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9682539682539683, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.953125, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04549052193760872, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9438202247191011, "eval_overall_precision": 0.9379652605459057, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2997, "eval_samples_per_second": 567.244, "eval_steps_per_second": 10.01, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.09890657663345337, "learning_rate": 4.000000000000001e-06, "loss": 0.0086, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04360693320631981, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9548872180451128, "eval_overall_precision": 0.9525, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2916, "eval_samples_per_second": 583.073, "eval_steps_per_second": 10.29, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.3869495689868927, "learning_rate": 3.5000000000000004e-06, "loss": 0.0079, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9682539682539683, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.953125, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04500889778137207, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.949748743718593, "eval_overall_precision": 0.949748743718593, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.291, "eval_samples_per_second": 584.268, "eval_steps_per_second": 10.311, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.626014769077301, "learning_rate": 3e-06, "loss": 0.0083, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04419349133968353, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.95, "eval_overall_precision": 0.945273631840796, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2875, "eval_samples_per_second": 591.343, "eval_steps_per_second": 10.435, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.140420064330101, "learning_rate": 2.5e-06, "loss": 0.0076, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04507080465555191, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2875, "eval_samples_per_second": 591.254, "eval_steps_per_second": 10.434, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.1293337047100067, "learning_rate": 2.0000000000000003e-06, "loss": 0.0072, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.9446808510638298, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9327731092436975, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9514563106796117, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9735099337748344, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04594406113028526, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9584905660377357, "eval_overall_precision": 0.9596977329974811, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2873, "eval_samples_per_second": 591.728, "eval_steps_per_second": 10.442, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.3379001021385193, "learning_rate": 1.5e-06, "loss": 0.0072, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.044891923666000366, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2886, "eval_samples_per_second": 589.09, "eval_steps_per_second": 10.396, "step": 9312 }, { "epoch": 98.0, "grad_norm": 1.1009929180145264, "learning_rate": 1.0000000000000002e-06, "loss": 0.0059, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04502749815583229, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2873, "eval_samples_per_second": 591.617, "eval_steps_per_second": 10.44, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.03712258115410805, "learning_rate": 5.000000000000001e-07, "loss": 0.0081, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04463572800159454, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2845, "eval_samples_per_second": 597.449, "eval_steps_per_second": 10.543, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.07331474125385284, "learning_rate": 0.0, "loss": 0.0074, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.044605545699596405, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2886, "eval_samples_per_second": 589.007, "eval_steps_per_second": 10.394, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 4293597283743744.0, "train_loss": 0.03720566373939316, "train_runtime": 529.4133, "train_samples_per_second": 289.188, "train_steps_per_second": 18.133 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 4293597283743744.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }