{ "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": 3.049309492111206, "learning_rate": 4.9500000000000004e-05, "loss": 0.2668, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.9184549356223175, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9145299145299145, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.9283489096573209, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9141104294478528, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03940283879637718, "eval_overall_accuracy": 0.9879220422728521, "eval_overall_f1": 0.9439601494396015, "eval_overall_precision": 0.9358024691358025, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2756, "eval_samples_per_second": 616.868, "eval_steps_per_second": 10.886, "step": 96 }, { "epoch": 2.0, "grad_norm": 1.6137810945510864, "learning_rate": 4.9e-05, "loss": 0.0634, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.932475884244373, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9477124183006536, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04595618322491646, "eval_overall_accuracy": 0.9881965413121054, "eval_overall_f1": 0.9496221662468514, "eval_overall_precision": 0.952020202020202, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2866, "eval_samples_per_second": 593.078, "eval_steps_per_second": 10.466, "step": 192 }, { "epoch": 3.0, "grad_norm": 1.8293578624725342, "learning_rate": 4.85e-05, "loss": 0.032, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.9391304347826087, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9473684210526315, "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": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04407869651913643, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.9534591194968555, "eval_overall_precision": 0.9546599496221663, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2736, "eval_samples_per_second": 621.343, "eval_steps_per_second": 10.965, "step": 288 }, { "epoch": 4.0, "grad_norm": 1.4507431983947754, "learning_rate": 4.8e-05, "loss": 0.022, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.9561403508771931, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9732142857142857, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9341692789968652, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9254658385093167, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.044241927564144135, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.957286432160804, "eval_overall_precision": 0.957286432160804, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2828, "eval_samples_per_second": 601.144, "eval_steps_per_second": 10.608, "step": 384 }, { "epoch": 5.0, "grad_norm": 2.4056336879730225, "learning_rate": 4.75e-05, "loss": 0.0143, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.9535864978902953, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9338842975206612, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04740007221698761, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9597989949748744, "eval_overall_precision": 0.9597989949748744, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2733, "eval_samples_per_second": 621.921, "eval_steps_per_second": 10.975, "step": 480 }, { "epoch": 6.0, "grad_norm": 0.2397124469280243, "learning_rate": 4.7e-05, "loss": 0.0122, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.9446808510638298, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9327731092436975, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.934640522875817, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9662162162162162, "eval_ORGANIZATION_recall": 0.9050632911392406, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.058103881776332855, "eval_overall_accuracy": 0.9884710403513588, "eval_overall_f1": 0.9518987341772152, "eval_overall_precision": 0.9591836734693877, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2739, "eval_samples_per_second": 620.573, "eval_steps_per_second": 10.951, "step": 576 }, { "epoch": 7.0, "grad_norm": 0.054269399493932724, "learning_rate": 4.6500000000000005e-05, "loss": 0.0062, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.9568965517241379, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9568965517241379, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "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": 0.05784749239683151, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9622166246851386, "eval_overall_precision": 0.9646464646464646, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2745, "eval_samples_per_second": 619.39, "eval_steps_per_second": 10.93, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.038999129086732864, "learning_rate": 4.600000000000001e-05, "loss": 0.007, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.9655172413793104, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9655172413793104, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9490445859872612, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9551282051282052, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.06084027141332626, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9660377358490567, "eval_overall_precision": 0.9672544080604534, "eval_overall_recall": 0.964824120603015, "eval_runtime": 0.2827, "eval_samples_per_second": 601.28, "eval_steps_per_second": 10.611, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.102794349193573, "learning_rate": 4.55e-05, "loss": 0.0049, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.9446808510638298, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9327731092436975, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9250814332247557, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9530201342281879, "eval_ORGANIZATION_recall": 0.8987341772151899, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.06557527184486389, "eval_overall_accuracy": 0.9873730441943454, "eval_overall_f1": 0.9506953223767383, "eval_overall_precision": 0.9567430025445293, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2897, "eval_samples_per_second": 586.732, "eval_steps_per_second": 10.354, "step": 864 }, { "epoch": 10.0, "grad_norm": 0.3394320011138916, "learning_rate": 4.5e-05, "loss": 0.0056, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.9568965517241379, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9568965517241379, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9363057324840764, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9423076923076923, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05663624778389931, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.957286432160804, "eval_overall_precision": 0.957286432160804, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2755, "eval_samples_per_second": 617.0, "eval_steps_per_second": 10.888, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.001915093045681715, "learning_rate": 4.4500000000000004e-05, "loss": 0.0046, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9196141479099678, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.934640522875817, "eval_ORGANIZATION_recall": 0.9050632911392406, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07088989019393921, "eval_overall_accuracy": 0.9879220422728521, "eval_overall_f1": 0.9521410579345088, "eval_overall_precision": 0.9545454545454546, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2767, "eval_samples_per_second": 614.422, "eval_steps_per_second": 10.843, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.015670381486415863, "learning_rate": 4.4000000000000006e-05, "loss": 0.0022, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "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": 0.07214923948049545, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9611041405269761, "eval_overall_precision": 0.9598997493734336, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2868, "eval_samples_per_second": 592.77, "eval_steps_per_second": 10.461, "step": 1152 }, { "epoch": 13.0, "grad_norm": 0.002585033653303981, "learning_rate": 4.35e-05, "loss": 0.0048, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.9527896995708154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9487179487179487, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.946031746031746, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9490445859872612, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05439727008342743, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9598997493734336, "eval_overall_precision": 0.9575, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2779, "eval_samples_per_second": 611.673, "eval_steps_per_second": 10.794, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.06117913872003555, "learning_rate": 4.3e-05, "loss": 0.0029, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.9565217391304347, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9649122807017544, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9463722397476341, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9433962264150944, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.06017274409532547, "eval_overall_accuracy": 0.9917650288223991, "eval_overall_f1": 0.9623115577889447, "eval_overall_precision": 0.9623115577889447, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2774, "eval_samples_per_second": 612.896, "eval_steps_per_second": 10.816, "step": 1344 }, { "epoch": 15.0, "grad_norm": 4.67892599105835, "learning_rate": 4.25e-05, "loss": 0.0031, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.9437229437229439, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9478260869565217, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.939297124600639, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9483870967741935, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.06778218597173691, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9546599496221662, "eval_overall_precision": 0.9570707070707071, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2794, "eval_samples_per_second": 608.342, "eval_steps_per_second": 10.735, "step": 1440 }, { "epoch": 16.0, "grad_norm": 0.5296499729156494, "learning_rate": 4.2e-05, "loss": 0.0039, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.935064935064935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.96, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.08199746906757355, "eval_overall_accuracy": 0.987098545155092, "eval_overall_f1": 0.9506953223767383, "eval_overall_precision": 0.9567430025445293, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2762, "eval_samples_per_second": 615.429, "eval_steps_per_second": 10.861, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.0023074958007782698, "learning_rate": 4.15e-05, "loss": 0.0021, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.9620253164556961, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9421487603305785, "eval_LOCATION_recall": 0.9827586206896551, "eval_ORGANIZATION_f1": 0.9514563106796117, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9735099337748344, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07931017875671387, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.964824120603015, "eval_overall_precision": 0.964824120603015, "eval_overall_recall": 0.964824120603015, "eval_runtime": 0.2743, "eval_samples_per_second": 619.839, "eval_steps_per_second": 10.938, "step": 1632 }, { "epoch": 18.0, "grad_norm": 1.223575472831726, "learning_rate": 4.1e-05, "loss": 0.0035, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.9310344827586207, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9310344827586207, "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": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08444110304117203, "eval_overall_accuracy": 0.9879220422728521, "eval_overall_f1": 0.950943396226415, "eval_overall_precision": 0.9521410579345088, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2749, "eval_samples_per_second": 618.352, "eval_steps_per_second": 10.912, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.030457496643066406, "learning_rate": 4.05e-05, "loss": 0.0039, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.9372384937238494, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9105691056910569, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9342105263157895, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9726027397260274, "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": 0.090733103454113, "eval_overall_accuracy": 0.9868240461158386, "eval_overall_f1": 0.9508196721311476, "eval_overall_precision": 0.9544303797468354, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2734, "eval_samples_per_second": 621.759, "eval_steps_per_second": 10.972, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.002068708650767803, "learning_rate": 4e-05, "loss": 0.0014, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.06293763220310211, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9610062893081761, "eval_overall_precision": 0.9622166246851386, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2744, "eval_samples_per_second": 619.493, "eval_steps_per_second": 10.932, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.05384279042482376, "learning_rate": 3.9500000000000005e-05, "loss": 0.0019, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9548387096774194, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9736842105263158, "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": 0.06547907739877701, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.964735516372796, "eval_overall_precision": 0.9671717171717171, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2814, "eval_samples_per_second": 604.105, "eval_steps_per_second": 10.661, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.06973671913146973, "learning_rate": 3.9000000000000006e-05, "loss": 0.0021, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.9527896995708154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9487179487179487, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9400630914826499, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9371069182389937, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.059321921318769455, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9586983729662076, "eval_overall_precision": 0.9551122194513716, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2794, "eval_samples_per_second": 608.393, "eval_steps_per_second": 10.736, "step": 2112 }, { "epoch": 23.0, "grad_norm": 0.0021264716051518917, "learning_rate": 3.85e-05, "loss": 0.0038, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "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": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.069790780544281, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.9559748427672956, "eval_overall_precision": 0.9571788413098237, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2838, "eval_samples_per_second": 598.989, "eval_steps_per_second": 10.57, "step": 2208 }, { "epoch": 24.0, "grad_norm": 0.0016756883123889565, "learning_rate": 3.8e-05, "loss": 0.0024, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "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": 0.06858450919389725, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.962121212121212, "eval_overall_precision": 0.9670050761421319, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2771, "eval_samples_per_second": 613.591, "eval_steps_per_second": 10.828, "step": 2304 }, { "epoch": 25.0, "grad_norm": 2.1548383235931396, "learning_rate": 3.7500000000000003e-05, "loss": 0.0032, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.9367088607594937, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9173553719008265, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9260450160771704, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9411764705882353, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0781761035323143, "eval_overall_accuracy": 0.9873730441943454, "eval_overall_f1": 0.9485570890840652, "eval_overall_precision": 0.9473684210526315, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2779, "eval_samples_per_second": 611.738, "eval_steps_per_second": 10.795, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.028054367750883102, "learning_rate": 3.7e-05, "loss": 0.0028, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.954248366013072, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9864864864864865, "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": 0.08411029726266861, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9582806573957017, "eval_overall_precision": 0.9643765903307888, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2758, "eval_samples_per_second": 616.493, "eval_steps_per_second": 10.879, "step": 2496 }, { "epoch": 27.0, "grad_norm": 0.059956323355436325, "learning_rate": 3.65e-05, "loss": 0.0024, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9367088607594937, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9173553719008265, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9523809523809524, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9554140127388535, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07615868002176285, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.958801498127341, "eval_overall_precision": 0.9528535980148883, "eval_overall_recall": 0.964824120603015, "eval_runtime": 0.2778, "eval_samples_per_second": 611.955, "eval_steps_per_second": 10.799, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.02544810064136982, "learning_rate": 3.6e-05, "loss": 0.0065, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9482758620689655, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9482758620689655, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.934640522875817, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9662162162162162, "eval_ORGANIZATION_recall": 0.9050632911392406, "eval_PERSON_f1": 0.9682539682539683, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.953125, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.09427817910909653, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.949367088607595, "eval_overall_precision": 0.9566326530612245, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2757, "eval_samples_per_second": 616.558, "eval_steps_per_second": 10.88, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.006958332844078541, "learning_rate": 3.55e-05, "loss": 0.0026, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9527896995708154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9487179487179487, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9381107491856678, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9664429530201343, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.09594732522964478, "eval_overall_accuracy": 0.9873730441943454, "eval_overall_f1": 0.9581749049429658, "eval_overall_precision": 0.9667519181585678, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2894, "eval_samples_per_second": 587.475, "eval_steps_per_second": 10.367, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.0011697375448420644, "learning_rate": 3.5e-05, "loss": 0.002, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "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": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07324973493814468, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9584905660377357, "eval_overall_precision": 0.9596977329974811, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2758, "eval_samples_per_second": 616.404, "eval_steps_per_second": 10.878, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.0018024586606770754, "learning_rate": 3.45e-05, "loss": 0.0012, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.9527896995708154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9487179487179487, "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": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08078110963106155, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9633375474083439, "eval_overall_precision": 0.9694656488549618, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2895, "eval_samples_per_second": 587.277, "eval_steps_per_second": 10.364, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.001483693951740861, "learning_rate": 3.4000000000000007e-05, "loss": 0.001, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9545454545454545, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.98, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08459720760583878, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.965865992414665, "eval_overall_precision": 0.9720101781170484, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2764, "eval_samples_per_second": 614.944, "eval_steps_per_second": 10.852, "step": 3072 }, { "epoch": 33.0, "grad_norm": 4.3353071212768555, "learning_rate": 3.35e-05, "loss": 0.0018, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.9527896995708154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9487179487179487, "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": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.09489604830741882, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9633375474083439, "eval_overall_precision": 0.9694656488549618, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2981, "eval_samples_per_second": 570.187, "eval_steps_per_second": 10.062, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.0007446123054251075, "learning_rate": 3.3e-05, "loss": 0.0012, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "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": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.09648539870977402, "eval_overall_accuracy": 0.9879220422728521, "eval_overall_f1": 0.9546599496221662, "eval_overall_precision": 0.9570707070707071, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2797, "eval_samples_per_second": 607.836, "eval_steps_per_second": 10.727, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.04847164824604988, "learning_rate": 3.2500000000000004e-05, "loss": 0.0025, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.9385964912280702, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9553571428571429, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.9367088607594937, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9367088607594937, "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": 0.10111605376005173, "eval_overall_accuracy": 0.9879220422728521, "eval_overall_f1": 0.9521410579345088, "eval_overall_precision": 0.9545454545454546, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2814, "eval_samples_per_second": 604.206, "eval_steps_per_second": 10.662, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.006004320923238993, "learning_rate": 3.2000000000000005e-05, "loss": 0.0029, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.9527896995708154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9487179487179487, "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": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.09128763526678085, "eval_overall_accuracy": 0.9881965413121054, "eval_overall_f1": 0.9584905660377357, "eval_overall_precision": 0.9596977329974811, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2762, "eval_samples_per_second": 615.538, "eval_steps_per_second": 10.862, "step": 3456 }, { "epoch": 37.0, "grad_norm": 0.5727871060371399, "learning_rate": 3.15e-05, "loss": 0.0037, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9430379746835443, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9430379746835443, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05428982526063919, "eval_overall_accuracy": 0.9923140269009059, "eval_overall_f1": 0.96, "eval_overall_precision": 0.9552238805970149, "eval_overall_recall": 0.964824120603015, "eval_runtime": 0.2785, "eval_samples_per_second": 610.43, "eval_steps_per_second": 10.772, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.0067481170408427715, "learning_rate": 3.1e-05, "loss": 0.002, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9527896995708154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9487179487179487, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9430379746835443, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9430379746835443, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.06546945124864578, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9598997493734336, "eval_overall_precision": 0.9575, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2747, "eval_samples_per_second": 618.844, "eval_steps_per_second": 10.921, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.0010467551182955503, "learning_rate": 3.05e-05, "loss": 0.0015, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.9523809523809523, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9565217391304348, "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": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0786169022321701, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9620253164556962, "eval_overall_precision": 0.9693877551020408, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2771, "eval_samples_per_second": 613.511, "eval_steps_per_second": 10.827, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.0016867171507328749, "learning_rate": 3e-05, "loss": 0.001, "step": 3840 }, { "epoch": 40.0, "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": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07218839973211288, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9584905660377357, "eval_overall_precision": 0.9596977329974811, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2825, "eval_samples_per_second": 601.73, "eval_steps_per_second": 10.619, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.02849876880645752, "learning_rate": 2.95e-05, "loss": 0.0021, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.9623430962343097, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9349593495934959, "eval_LOCATION_recall": 0.9913793103448276, "eval_ORGANIZATION_f1": 0.9548387096774194, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9736842105263158, "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": 0.07220359891653061, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9674185463659147, "eval_overall_precision": 0.965, "eval_overall_recall": 0.9698492462311558, "eval_runtime": 0.2929, "eval_samples_per_second": 580.454, "eval_steps_per_second": 10.243, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.010168996639549732, "learning_rate": 2.9e-05, "loss": 0.0018, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.9482758620689655, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9482758620689655, "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": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07643670588731766, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9583858764186632, "eval_overall_precision": 0.9620253164556962, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2741, "eval_samples_per_second": 620.184, "eval_steps_per_second": 10.944, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.0018144345376640558, "learning_rate": 2.8499999999999998e-05, "loss": 0.0009, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9514563106796117, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9735099337748344, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.08542946726083755, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9633375474083439, "eval_overall_precision": 0.9694656488549618, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.278, "eval_samples_per_second": 611.507, "eval_steps_per_second": 10.791, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.00032525003189221025, "learning_rate": 2.8000000000000003e-05, "loss": 0.0007, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9514563106796117, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9735099337748344, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.07778704911470413, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.962121212121212, "eval_overall_precision": 0.9670050761421319, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2764, "eval_samples_per_second": 614.952, "eval_steps_per_second": 10.852, "step": 4224 }, { "epoch": 45.0, "grad_norm": 8.090188026428223, "learning_rate": 2.7500000000000004e-05, "loss": 0.0018, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.943231441048035, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9557522123893806, "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": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.08798850327730179, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9518987341772152, "eval_overall_precision": 0.9591836734693877, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2772, "eval_samples_per_second": 613.239, "eval_steps_per_second": 10.822, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.06047314405441284, "learning_rate": 2.7000000000000002e-05, "loss": 0.0022, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.961038961038961, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9866666666666667, "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": 0.08229512721300125, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9646464646464646, "eval_overall_precision": 0.9695431472081218, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2747, "eval_samples_per_second": 618.874, "eval_steps_per_second": 10.921, "step": 4416 }, { "epoch": 47.0, "grad_norm": 11.46757698059082, "learning_rate": 2.6500000000000004e-05, "loss": 0.0013, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.9482758620689655, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9482758620689655, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9545454545454545, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.98, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.09129085391759872, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9620253164556962, "eval_overall_precision": 0.9693877551020408, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.275, "eval_samples_per_second": 618.168, "eval_steps_per_second": 10.909, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.003439373802393675, "learning_rate": 2.6000000000000002e-05, "loss": 0.0013, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9576271186440678, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9416666666666667, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9579288025889968, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9801324503311258, "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": 0.08189716935157776, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9672544080604534, "eval_overall_precision": 0.9696969696969697, "eval_overall_recall": 0.964824120603015, "eval_runtime": 0.2766, "eval_samples_per_second": 614.694, "eval_steps_per_second": 10.848, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.0003379171248525381, "learning_rate": 2.5500000000000003e-05, "loss": 0.0005, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9548387096774194, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9736842105263158, "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": 0.0734885111451149, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.964735516372796, "eval_overall_precision": 0.9671717171717171, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2755, "eval_samples_per_second": 617.144, "eval_steps_per_second": 10.891, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.0015403638826683164, "learning_rate": 2.5e-05, "loss": 0.0011, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.9482758620689655, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9482758620689655, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.939297124600639, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9483870967741935, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07724250108003616, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9571788413098236, "eval_overall_precision": 0.9595959595959596, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2751, "eval_samples_per_second": 618.052, "eval_steps_per_second": 10.907, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.005287709180265665, "learning_rate": 2.45e-05, "loss": 0.0021, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9514563106796117, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9735099337748344, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08125069737434387, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9646464646464646, "eval_overall_precision": 0.9695431472081218, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2767, "eval_samples_per_second": 614.409, "eval_steps_per_second": 10.843, "step": 4896 }, { "epoch": 52.0, "grad_norm": 0.055303167551755905, "learning_rate": 2.4e-05, "loss": 0.0006, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9658119658119658, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9576271186440678, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9545454545454545, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.98, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.09267558157444, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9683944374209861, "eval_overall_precision": 0.9745547073791349, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.273, "eval_samples_per_second": 622.619, "eval_steps_per_second": 10.987, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.0012606492964550853, "learning_rate": 2.35e-05, "loss": 0.0007, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9617021276595743, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9495798319327731, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9615384615384615, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.974025974025974, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07907352596521378, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9698492462311558, "eval_overall_precision": 0.9698492462311558, "eval_overall_recall": 0.9698492462311558, "eval_runtime": 0.2759, "eval_samples_per_second": 616.184, "eval_steps_per_second": 10.874, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.0005766572430729866, "learning_rate": 2.3000000000000003e-05, "loss": 0.0011, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.9617021276595743, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9495798319327731, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9617834394904459, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.967948717948718, "eval_ORGANIZATION_recall": 0.9556962025316456, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07217950373888016, "eval_overall_accuracy": 0.9928630249794126, "eval_overall_f1": 0.9699248120300752, "eval_overall_precision": 0.9675, "eval_overall_recall": 0.9723618090452262, "eval_runtime": 0.2765, "eval_samples_per_second": 614.717, "eval_steps_per_second": 10.848, "step": 5184 }, { "epoch": 55.0, "grad_norm": 0.000538014282938093, "learning_rate": 2.25e-05, "loss": 0.0005, "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": 0.9556962025316456, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9556962025316456, "eval_ORGANIZATION_recall": 0.9556962025316456, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.07206634432077408, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9612015018773465, "eval_overall_precision": 0.9576059850374065, "eval_overall_recall": 0.964824120603015, "eval_runtime": 0.2784, "eval_samples_per_second": 610.702, "eval_steps_per_second": 10.777, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.00032307393848896027, "learning_rate": 2.2000000000000003e-05, "loss": 0.0005, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9617021276595743, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9495798319327731, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9712460063897763, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9806451612903225, "eval_ORGANIZATION_recall": 0.9620253164556962, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.07047928124666214, "eval_overall_accuracy": 0.993137524018666, "eval_overall_f1": 0.9723618090452262, "eval_overall_precision": 0.9723618090452262, "eval_overall_recall": 0.9723618090452262, "eval_runtime": 0.2765, "eval_samples_per_second": 614.867, "eval_steps_per_second": 10.851, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.43909594416618347, "learning_rate": 2.15e-05, "loss": 0.0003, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.9527896995708154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9487179487179487, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9584664536741213, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.967741935483871, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.06509443372488022, "eval_overall_accuracy": 0.9923140269009059, "eval_overall_f1": 0.964735516372796, "eval_overall_precision": 0.9671717171717171, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2841, "eval_samples_per_second": 598.449, "eval_steps_per_second": 10.561, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.020889485254883766, "learning_rate": 2.1e-05, "loss": 0.0011, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9568965517241379, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9568965517241379, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9617834394904459, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.967948717948718, "eval_ORGANIZATION_recall": 0.9556962025316456, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.07538726180791855, "eval_overall_accuracy": 0.9928630249794126, "eval_overall_f1": 0.9672544080604534, "eval_overall_precision": 0.9696969696969697, "eval_overall_recall": 0.964824120603015, "eval_runtime": 0.2767, "eval_samples_per_second": 614.294, "eval_steps_per_second": 10.84, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.00022353450185619295, "learning_rate": 2.05e-05, "loss": 0.0006, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9396551724137931, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9396551724137931, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9587301587301587, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9617834394904459, "eval_ORGANIZATION_recall": 0.9556962025316456, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.07177454978227615, "eval_overall_accuracy": 0.9923140269009059, "eval_overall_f1": 0.9610062893081761, "eval_overall_precision": 0.9622166246851386, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2865, "eval_samples_per_second": 593.345, "eval_steps_per_second": 10.471, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.00033137862919829786, "learning_rate": 2e-05, "loss": 0.0005, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.9617021276595743, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9495798319327731, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9545454545454545, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.98, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08699022233486176, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9671717171717171, "eval_overall_precision": 0.9720812182741116, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2736, "eval_samples_per_second": 621.399, "eval_steps_per_second": 10.966, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.0024979726877063513, "learning_rate": 1.9500000000000003e-05, "loss": 0.0004, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.9391304347826087, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9473684210526315, "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": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0686868205666542, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.958594730238394, "eval_overall_precision": 0.9573934837092731, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.28, "eval_samples_per_second": 607.233, "eval_steps_per_second": 10.716, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.0027951186057180166, "learning_rate": 1.9e-05, "loss": 0.0002, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "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": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.0982985645532608, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9593908629441623, "eval_overall_precision": 0.9692307692307692, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2763, "eval_samples_per_second": 615.344, "eval_steps_per_second": 10.859, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.0001827309897635132, "learning_rate": 1.85e-05, "loss": 0.0006, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08179256319999695, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9646464646464646, "eval_overall_precision": 0.9695431472081218, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2898, "eval_samples_per_second": 586.599, "eval_steps_per_second": 10.352, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.0003153543220832944, "learning_rate": 1.8e-05, "loss": 0.0002, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08577439934015274, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2729, "eval_samples_per_second": 622.862, "eval_steps_per_second": 10.992, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.000563719542697072, "learning_rate": 1.75e-05, "loss": 0.0005, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.9568965517241379, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9568965517241379, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9517684887459807, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9673202614379085, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.08837004005908966, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9609079445145018, "eval_overall_precision": 0.9645569620253165, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.275, "eval_samples_per_second": 618.13, "eval_steps_per_second": 10.908, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.001452375203371048, "learning_rate": 1.7000000000000003e-05, "loss": 0.001, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9527896995708154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9487179487179487, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9389067524115756, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.954248366013072, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.0770520567893982, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9546599496221662, "eval_overall_precision": 0.9570707070707071, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2871, "eval_samples_per_second": 592.039, "eval_steps_per_second": 10.448, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.0012776756193488836, "learning_rate": 1.65e-05, "loss": 0.0006, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9527896995708154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9487179487179487, "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": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08081617951393127, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9633375474083439, "eval_overall_precision": 0.9694656488549618, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2769, "eval_samples_per_second": 614.016, "eval_steps_per_second": 10.836, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.1566210240125656, "learning_rate": 1.6000000000000003e-05, "loss": 0.0002, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "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": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.074882373213768, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.964735516372796, "eval_overall_precision": 0.9671717171717171, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2768, "eval_samples_per_second": 614.249, "eval_steps_per_second": 10.84, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.000284359062789008, "learning_rate": 1.55e-05, "loss": 0.0011, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07836616784334183, "eval_overall_accuracy": 0.9917650288223991, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2749, "eval_samples_per_second": 618.479, "eval_steps_per_second": 10.914, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.0007131195743568242, "learning_rate": 1.5e-05, "loss": 0.0005, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07496281713247299, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9646464646464646, "eval_overall_precision": 0.9695431472081218, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2777, "eval_samples_per_second": 612.098, "eval_steps_per_second": 10.802, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.0003653345920611173, "learning_rate": 1.45e-05, "loss": 0.0001, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07584992796182632, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9646464646464646, "eval_overall_precision": 0.9695431472081218, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2767, "eval_samples_per_second": 614.357, "eval_steps_per_second": 10.842, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.00019452819833531976, "learning_rate": 1.4000000000000001e-05, "loss": 0.0005, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07706008106470108, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9646464646464646, "eval_overall_precision": 0.9695431472081218, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2789, "eval_samples_per_second": 609.626, "eval_steps_per_second": 10.758, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.00044385006185621023, "learning_rate": 1.3500000000000001e-05, "loss": 0.0004, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9389067524115756, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.954248366013072, "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": 0.07329727709293365, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9584905660377357, "eval_overall_precision": 0.9596977329974811, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2761, "eval_samples_per_second": 615.718, "eval_steps_per_second": 10.866, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.0001623950811335817, "learning_rate": 1.3000000000000001e-05, "loss": 0.0001, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9389067524115756, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.954248366013072, "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": 0.07398699223995209, "eval_overall_accuracy": 0.9917650288223991, "eval_overall_f1": 0.9596977329974811, "eval_overall_precision": 0.9621212121212122, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2763, "eval_samples_per_second": 615.183, "eval_steps_per_second": 10.856, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.00019397769938223064, "learning_rate": 1.25e-05, "loss": 0.0001, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "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": 0.07948914915323257, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.962121212121212, "eval_overall_precision": 0.9670050761421319, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.288, "eval_samples_per_second": 590.181, "eval_steps_per_second": 10.415, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.00020719191525131464, "learning_rate": 1.2e-05, "loss": 0.0002, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "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": 0.08004683256149292, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.962121212121212, "eval_overall_precision": 0.9670050761421319, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2748, "eval_samples_per_second": 618.639, "eval_steps_per_second": 10.917, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.0001302505552303046, "learning_rate": 1.1500000000000002e-05, "loss": 0.0002, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.9568965517241379, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9568965517241379, "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": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07812328636646271, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.962121212121212, "eval_overall_precision": 0.9670050761421319, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2724, "eval_samples_per_second": 624.165, "eval_steps_per_second": 11.015, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.00012877046538051218, "learning_rate": 1.1000000000000001e-05, "loss": 0.0002, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9514563106796117, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9735099337748344, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07981810718774796, "eval_overall_accuracy": 0.9917650288223991, "eval_overall_f1": 0.9646464646464646, "eval_overall_precision": 0.9695431472081218, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2748, "eval_samples_per_second": 618.678, "eval_steps_per_second": 10.918, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.00036683276994153857, "learning_rate": 1.05e-05, "loss": 0.0002, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9548387096774194, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9736842105263158, "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": 0.07850603759288788, "eval_overall_accuracy": 0.9925885259401592, "eval_overall_f1": 0.9671717171717171, "eval_overall_precision": 0.9720812182741116, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2871, "eval_samples_per_second": 592.19, "eval_steps_per_second": 10.45, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.00012351616169326007, "learning_rate": 1e-05, "loss": 0.0001, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9548387096774194, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9736842105263158, "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": 0.07939465343952179, "eval_overall_accuracy": 0.9925885259401592, "eval_overall_f1": 0.9671717171717171, "eval_overall_precision": 0.9720812182741116, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2831, "eval_samples_per_second": 600.495, "eval_steps_per_second": 10.597, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.00012122365296818316, "learning_rate": 9.5e-06, "loss": 0.0004, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9548387096774194, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9736842105263158, "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": 0.08115226775407791, "eval_overall_accuracy": 0.9925885259401592, "eval_overall_f1": 0.9671717171717171, "eval_overall_precision": 0.9720812182741116, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2786, "eval_samples_per_second": 610.221, "eval_steps_per_second": 10.769, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.00014256218855734915, "learning_rate": 9e-06, "loss": 0.0001, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "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": 0.08801557123661041, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.962121212121212, "eval_overall_precision": 0.9670050761421319, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2757, "eval_samples_per_second": 616.648, "eval_steps_per_second": 10.882, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.00044690616778098047, "learning_rate": 8.500000000000002e-06, "loss": 0.0001, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9658119658119658, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9576271186440678, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08324732631444931, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.965952080706179, "eval_overall_precision": 0.9696202531645569, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.278, "eval_samples_per_second": 611.537, "eval_steps_per_second": 10.792, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.00016260806296486408, "learning_rate": 8.000000000000001e-06, "loss": 0.0007, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "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": 0.08543501794338226, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.962121212121212, "eval_overall_precision": 0.9670050761421319, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2754, "eval_samples_per_second": 617.266, "eval_steps_per_second": 10.893, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.00022077209723647684, "learning_rate": 7.5e-06, "loss": 0.0001, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "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": 0.086331307888031, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.962121212121212, "eval_overall_precision": 0.9670050761421319, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2842, "eval_samples_per_second": 598.232, "eval_steps_per_second": 10.557, "step": 8160 }, { "epoch": 86.0, "grad_norm": 0.00011886875290656462, "learning_rate": 7.000000000000001e-06, "loss": 0.0001, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9572649572649573, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9491525423728814, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "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": 0.0853876993060112, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.962121212121212, "eval_overall_precision": 0.9670050761421319, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2739, "eval_samples_per_second": 620.608, "eval_steps_per_second": 10.952, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.00010190217290073633, "learning_rate": 6.5000000000000004e-06, "loss": 0.0001, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9517684887459807, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9673202614379085, "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": 0.07888607680797577, "eval_overall_accuracy": 0.9923140269009059, "eval_overall_f1": 0.965952080706179, "eval_overall_precision": 0.9696202531645569, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.276, "eval_samples_per_second": 616.031, "eval_steps_per_second": 10.871, "step": 8352 }, { "epoch": 88.0, "grad_norm": 8.675308345118538e-05, "learning_rate": 6e-06, "loss": 0.0001, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9699570815450644, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9658119658119658, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.07756155729293823, "eval_overall_accuracy": 0.9923140269009059, "eval_overall_f1": 0.965952080706179, "eval_overall_precision": 0.9696202531645569, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2738, "eval_samples_per_second": 620.93, "eval_steps_per_second": 10.958, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.16685304045677185, "learning_rate": 5.500000000000001e-06, "loss": 0.0002, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9568965517241379, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9568965517241379, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9487179487179488, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.961038961038961, "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": 0.07858122885227203, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2787, "eval_samples_per_second": 609.957, "eval_steps_per_second": 10.764, "step": 8544 }, { "epoch": 90.0, "grad_norm": 9.697148198029026e-05, "learning_rate": 5e-06, "loss": 0.0001, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9568965517241379, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9568965517241379, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9487179487179488, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.961038961038961, "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": 0.07982868701219559, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2747, "eval_samples_per_second": 618.797, "eval_steps_per_second": 10.92, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.00033092033118009567, "learning_rate": 4.5e-06, "loss": 0.0001, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08155547082424164, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2787, "eval_samples_per_second": 609.866, "eval_steps_per_second": 10.762, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.00010464302613399923, "learning_rate": 4.000000000000001e-06, "loss": 0.0005, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08192423731088638, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2824, "eval_samples_per_second": 602.029, "eval_steps_per_second": 10.624, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.002409059088677168, "learning_rate": 3.5000000000000004e-06, "loss": 0.0003, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08189983665943146, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2756, "eval_samples_per_second": 616.941, "eval_steps_per_second": 10.887, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.16550596058368683, "learning_rate": 3e-06, "loss": 0.0003, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08141722530126572, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2729, "eval_samples_per_second": 622.955, "eval_steps_per_second": 10.993, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.0001592171611264348, "learning_rate": 2.5e-06, "loss": 0.0001, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08141204714775085, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2846, "eval_samples_per_second": 597.272, "eval_steps_per_second": 10.54, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.00010054935410153121, "learning_rate": 2.0000000000000003e-06, "loss": 0.0001, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08156644552946091, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2743, "eval_samples_per_second": 619.773, "eval_steps_per_second": 10.937, "step": 9216 }, { "epoch": 97.0, "grad_norm": 7.648551400052384e-05, "learning_rate": 1.5e-06, "loss": 0.0001, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08174903690814972, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2863, "eval_samples_per_second": 593.712, "eval_steps_per_second": 10.477, "step": 9312 }, { "epoch": 98.0, "grad_norm": 0.0001711288350634277, "learning_rate": 1.0000000000000002e-06, "loss": 0.0001, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08208905160427094, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2744, "eval_samples_per_second": 619.524, "eval_steps_per_second": 10.933, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.00011955283116549253, "learning_rate": 5.000000000000001e-07, "loss": 0.0001, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08216139674186707, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2911, "eval_samples_per_second": 584.081, "eval_steps_per_second": 10.307, "step": 9504 }, { "epoch": 100.0, "grad_norm": 7.644743891432881e-05, "learning_rate": 0.0, "loss": 0.0001, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08217138797044754, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9634300126103404, "eval_overall_precision": 0.9670886075949368, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2751, "eval_samples_per_second": 618.039, "eval_steps_per_second": 10.907, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 3848029980850176.0, "train_loss": 0.005479360639195269, "train_runtime": 897.0572, "train_samples_per_second": 170.669, "train_steps_per_second": 10.702 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 3848029980850176.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }