{ "best_metric": null, "best_model_checkpoint": null, "epoch": 100.0, "eval_steps": 500, "global_step": 9600, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 1.143059253692627, "learning_rate": 4.9500000000000004e-05, "loss": 0.8464, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.0, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.0, "eval_LOCATION_recall": 0.0, "eval_ORGANIZATION_f1": 0.2663316582914573, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.2409090909090909, "eval_ORGANIZATION_recall": 0.29775280898876405, "eval_PERSON_f1": 0.33935018050541516, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.31543624161073824, "eval_PERSON_recall": 0.3671875, "eval_loss": 0.3799312710762024, "eval_overall_accuracy": 0.8747638326585695, "eval_overall_f1": 0.2607561929595828, "eval_overall_precision": 0.26666666666666666, "eval_overall_recall": 0.25510204081632654, "eval_runtime": 0.2875, "eval_samples_per_second": 591.375, "eval_steps_per_second": 10.436, "step": 96 }, { "epoch": 2.0, "grad_norm": 1.1125810146331787, "learning_rate": 4.9e-05, "loss": 0.3634, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.4, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.3508771929824561, "eval_LOCATION_recall": 0.46511627906976744, "eval_ORGANIZATION_f1": 0.5941176470588235, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.6234567901234568, "eval_ORGANIZATION_recall": 0.5674157303370787, "eval_PERSON_f1": 0.6598639455782312, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.5843373493975904, "eval_PERSON_recall": 0.7578125, "eval_loss": 0.21567517518997192, "eval_overall_accuracy": 0.9330634278002699, "eval_overall_f1": 0.5707434052757793, "eval_overall_precision": 0.5384615384615384, "eval_overall_recall": 0.6071428571428571, "eval_runtime": 0.2882, "eval_samples_per_second": 589.783, "eval_steps_per_second": 10.408, "step": 192 }, { "epoch": 3.0, "grad_norm": 0.8366958498954773, "learning_rate": 4.85e-05, "loss": 0.1975, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.7692307692307692, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.7831325301204819, "eval_LOCATION_recall": 0.7558139534883721, "eval_ORGANIZATION_f1": 0.7801047120418848, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.7303921568627451, "eval_ORGANIZATION_recall": 0.8370786516853933, "eval_PERSON_f1": 0.9615384615384616, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.946969696969697, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.09421665221452713, "eval_overall_accuracy": 0.9692307692307692, "eval_overall_f1": 0.8360049321824907, "eval_overall_precision": 0.8090692124105012, "eval_overall_recall": 0.8647959183673469, "eval_runtime": 0.2893, "eval_samples_per_second": 587.54, "eval_steps_per_second": 10.368, "step": 288 }, { "epoch": 4.0, "grad_norm": 0.7830145955085754, "learning_rate": 4.8e-05, "loss": 0.1277, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.7802197802197802, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.7395833333333334, "eval_LOCATION_recall": 0.8255813953488372, "eval_ORGANIZATION_f1": 0.7777777777777778, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.8109756097560976, "eval_ORGANIZATION_recall": 0.7471910112359551, "eval_PERSON_f1": 0.9575289575289575, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9465648854961832, "eval_PERSON_recall": 0.96875, "eval_loss": 0.09090130031108856, "eval_overall_accuracy": 0.970310391363023, "eval_overall_f1": 0.8378033205619412, "eval_overall_precision": 0.8388746803069054, "eval_overall_recall": 0.8367346938775511, "eval_runtime": 0.2987, "eval_samples_per_second": 569.056, "eval_steps_per_second": 10.042, "step": 384 }, { "epoch": 5.0, "grad_norm": 0.5714436769485474, "learning_rate": 4.75e-05, "loss": 0.1039, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.8043478260869564, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.7551020408163265, "eval_LOCATION_recall": 0.8604651162790697, "eval_ORGANIZATION_f1": 0.8876404494382022, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.8876404494382022, "eval_ORGANIZATION_recall": 0.8876404494382022, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.06319117546081543, "eval_overall_accuracy": 0.9784075573549258, "eval_overall_f1": 0.8944723618090452, "eval_overall_precision": 0.8811881188118812, "eval_overall_recall": 0.9081632653061225, "eval_runtime": 0.2924, "eval_samples_per_second": 581.478, "eval_steps_per_second": 10.261, "step": 480 }, { "epoch": 6.0, "grad_norm": 0.9416435956954956, "learning_rate": 4.7e-05, "loss": 0.0916, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.7722772277227723, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.6724137931034483, "eval_LOCATION_recall": 0.9069767441860465, "eval_ORGANIZATION_f1": 0.8408408408408408, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9032258064516129, "eval_ORGANIZATION_recall": 0.7865168539325843, "eval_PERSON_f1": 0.9689922480620154, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9615384615384616, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.07187766581773758, "eval_overall_accuracy": 0.9743589743589743, "eval_overall_f1": 0.8650693568726356, "eval_overall_precision": 0.8553615960099751, "eval_overall_recall": 0.875, "eval_runtime": 0.2897, "eval_samples_per_second": 586.733, "eval_steps_per_second": 10.354, "step": 576 }, { "epoch": 7.0, "grad_norm": 0.4728829860687256, "learning_rate": 4.6500000000000005e-05, "loss": 0.0815, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.8505747126436781, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8409090909090909, "eval_LOCATION_recall": 0.8604651162790697, "eval_ORGANIZATION_f1": 0.9051490514905149, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.8743455497382199, "eval_ORGANIZATION_recall": 0.9382022471910112, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.052802398800849915, "eval_overall_accuracy": 0.9840755735492578, "eval_overall_f1": 0.9172932330827067, "eval_overall_precision": 0.9014778325123153, "eval_overall_recall": 0.9336734693877551, "eval_runtime": 0.2893, "eval_samples_per_second": 587.658, "eval_steps_per_second": 10.37, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.4887949526309967, "learning_rate": 4.600000000000001e-05, "loss": 0.0776, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.8771929824561403, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.872093023255814, "eval_ORGANIZATION_f1": 0.9116022099447514, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.8967391304347826, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9725490196078432, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9763779527559056, "eval_PERSON_recall": 0.96875, "eval_loss": 0.05136692896485329, "eval_overall_accuracy": 0.982995951417004, "eval_overall_f1": 0.9238578680203046, "eval_overall_precision": 0.9191919191919192, "eval_overall_recall": 0.9285714285714286, "eval_runtime": 0.2896, "eval_samples_per_second": 587.103, "eval_steps_per_second": 10.361, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.5312128067016602, "learning_rate": 4.55e-05, "loss": 0.0697, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.8677248677248678, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.7961165048543689, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.8914285714285715, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9069767441860465, "eval_ORGANIZATION_recall": 0.8764044943820225, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.051918163895606995, "eval_overall_accuracy": 0.9824561403508771, "eval_overall_f1": 0.9143576826196474, "eval_overall_precision": 0.9029850746268657, "eval_overall_recall": 0.9260204081632653, "eval_runtime": 0.291, "eval_samples_per_second": 584.135, "eval_steps_per_second": 10.308, "step": 864 }, { "epoch": 10.0, "grad_norm": 0.7800296545028687, "learning_rate": 4.5e-05, "loss": 0.0641, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.8770053475935828, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8118811881188119, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.8945868945868946, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9075144508670521, "eval_ORGANIZATION_recall": 0.8820224719101124, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05729928985238075, "eval_overall_accuracy": 0.9802968960863697, "eval_overall_f1": 0.9168765743073048, "eval_overall_precision": 0.9054726368159204, "eval_overall_recall": 0.9285714285714286, "eval_runtime": 0.2894, "eval_samples_per_second": 587.422, "eval_steps_per_second": 10.366, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.5551481246948242, "learning_rate": 4.4500000000000004e-05, "loss": 0.0613, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9279538904899135, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9526627218934911, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.049437254667282104, "eval_overall_accuracy": 0.9854251012145749, "eval_overall_f1": 0.940127388535032, "eval_overall_precision": 0.9389312977099237, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2886, "eval_samples_per_second": 588.966, "eval_steps_per_second": 10.394, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.44838830828666687, "learning_rate": 4.4000000000000006e-05, "loss": 0.0578, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.8901098901098902, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.84375, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9055555555555556, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.8956043956043956, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9727626459143969, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9689922480620154, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04607177525758743, "eval_overall_accuracy": 0.9856950067476383, "eval_overall_f1": 0.9236545682102628, "eval_overall_precision": 0.9066339066339066, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2882, "eval_samples_per_second": 589.933, "eval_steps_per_second": 10.411, "step": 1152 }, { "epoch": 13.0, "grad_norm": 0.8234059810638428, "learning_rate": 4.35e-05, "loss": 0.0527, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.8913043478260869, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8367346938775511, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9080459770114941, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9294117647058824, "eval_ORGANIZATION_recall": 0.8876404494382022, "eval_PERSON_f1": 0.9612403100775193, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9538461538461539, "eval_PERSON_recall": 0.96875, "eval_loss": 0.058097951114177704, "eval_overall_accuracy": 0.9827260458839406, "eval_overall_f1": 0.9215189873417722, "eval_overall_precision": 0.914572864321608, "eval_overall_recall": 0.9285714285714286, "eval_runtime": 0.2908, "eval_samples_per_second": 584.659, "eval_steps_per_second": 10.318, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.21934597194194794, "learning_rate": 4.3e-05, "loss": 0.0485, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.8666666666666666, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8297872340425532, "eval_LOCATION_recall": 0.9069767441860465, "eval_ORGANIZATION_f1": 0.9173789173789175, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.930635838150289, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9689922480620154, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9615384615384616, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05226796120405197, "eval_overall_accuracy": 0.982995951417004, "eval_overall_f1": 0.9226869455006337, "eval_overall_precision": 0.9168765743073047, "eval_overall_recall": 0.9285714285714286, "eval_runtime": 0.2895, "eval_samples_per_second": 587.264, "eval_steps_per_second": 10.363, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.8115574717521667, "learning_rate": 4.25e-05, "loss": 0.0467, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.8977272727272728, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8777777777777778, "eval_LOCATION_recall": 0.9186046511627907, "eval_ORGANIZATION_f1": 0.9239436619718311, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9265536723163842, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9689922480620154, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9615384615384616, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.0490017831325531, "eval_overall_accuracy": 0.9848852901484481, "eval_overall_f1": 0.9328263624841572, "eval_overall_precision": 0.9269521410579346, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.2927, "eval_samples_per_second": 580.716, "eval_steps_per_second": 10.248, "step": 1440 }, { "epoch": 16.0, "grad_norm": 1.156256914138794, "learning_rate": 4.2e-05, "loss": 0.0459, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.8839779005524862, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8421052631578947, "eval_LOCATION_recall": 0.9302325581395349, "eval_ORGANIZATION_f1": 0.92090395480226, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9261363636363636, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9649805447470817, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9612403100775194, "eval_PERSON_recall": 0.96875, "eval_loss": 0.046931665390729904, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9267676767676768, "eval_overall_precision": 0.9175, "eval_overall_recall": 0.9362244897959183, "eval_runtime": 0.2894, "eval_samples_per_second": 587.341, "eval_steps_per_second": 10.365, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.4832126200199127, "learning_rate": 4.15e-05, "loss": 0.0449, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9126760563380282, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9152542372881356, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9689922480620154, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9615384615384616, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.048053063452243805, "eval_overall_accuracy": 0.9854251012145749, "eval_overall_f1": 0.929471032745592, "eval_overall_precision": 0.917910447761194, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2919, "eval_samples_per_second": 582.34, "eval_steps_per_second": 10.277, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.5611772537231445, "learning_rate": 4.1e-05, "loss": 0.0419, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.888888888888889, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.851063829787234, "eval_LOCATION_recall": 0.9302325581395349, "eval_ORGANIZATION_f1": 0.9132947976878611, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9404761904761905, "eval_ORGANIZATION_recall": 0.8876404494382022, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.051899902522563934, "eval_overall_accuracy": 0.9843454790823212, "eval_overall_f1": 0.928388746803069, "eval_overall_precision": 0.9307692307692308, "eval_overall_recall": 0.9260204081632653, "eval_runtime": 0.2872, "eval_samples_per_second": 591.978, "eval_steps_per_second": 10.447, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.35904592275619507, "learning_rate": 4.05e-05, "loss": 0.0408, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.9132947976878613, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9080459770114943, "eval_LOCATION_recall": 0.9186046511627907, "eval_ORGANIZATION_f1": 0.9444444444444444, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9340659340659341, "eval_ORGANIZATION_recall": 0.9550561797752809, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.040789972990751266, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9480354879594423, "eval_overall_precision": 0.9420654911838791, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.2833, "eval_samples_per_second": 600.027, "eval_steps_per_second": 10.589, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.4714739918708801, "learning_rate": 4e-05, "loss": 0.0378, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.8648648648648648, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8080808080808081, "eval_LOCATION_recall": 0.9302325581395349, "eval_ORGANIZATION_f1": 0.9230769230769231, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9364161849710982, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.0570121705532074, "eval_overall_accuracy": 0.9851551956815114, "eval_overall_f1": 0.9267676767676768, "eval_overall_precision": 0.9175, "eval_overall_recall": 0.9362244897959183, "eval_runtime": 0.2939, "eval_samples_per_second": 578.495, "eval_steps_per_second": 10.209, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.9169831275939941, "learning_rate": 3.9500000000000005e-05, "loss": 0.0373, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9217391304347825, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9520958083832335, "eval_ORGANIZATION_recall": 0.8932584269662921, "eval_PERSON_f1": 0.9727626459143969, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9689922480620154, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.0537736676633358, "eval_overall_accuracy": 0.9846153846153847, "eval_overall_f1": 0.9348659003831418, "eval_overall_precision": 0.9360613810741688, "eval_overall_recall": 0.9336734693877551, "eval_runtime": 0.2912, "eval_samples_per_second": 583.82, "eval_steps_per_second": 10.303, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.6727938055992126, "learning_rate": 3.9000000000000006e-05, "loss": 0.0369, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.8852459016393444, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8350515463917526, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9230769230769231, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9364161849710982, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9727626459143969, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9689922480620154, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05055347830057144, "eval_overall_accuracy": 0.9854251012145749, "eval_overall_f1": 0.9304677623261693, "eval_overall_precision": 0.9223057644110275, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.2852, "eval_samples_per_second": 596.115, "eval_steps_per_second": 10.52, "step": 2112 }, { "epoch": 23.0, "grad_norm": 0.7414480447769165, "learning_rate": 3.85e-05, "loss": 0.0357, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9344729344729344, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9479768786127167, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9727626459143969, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9689922480620154, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.0469173826277256, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9417721518987343, "eval_overall_precision": 0.9346733668341709, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2847, "eval_samples_per_second": 597.072, "eval_steps_per_second": 10.537, "step": 2208 }, { "epoch": 24.0, "grad_norm": 1.0000765323638916, "learning_rate": 3.8e-05, "loss": 0.0347, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9295774647887324, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9322033898305084, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.049894943833351135, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.9417721518987343, "eval_overall_precision": 0.9346733668341709, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2944, "eval_samples_per_second": 577.418, "eval_steps_per_second": 10.19, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.3176489472389221, "learning_rate": 3.7500000000000003e-05, "loss": 0.034, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.898876404494382, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8695652173913043, "eval_LOCATION_recall": 0.9302325581395349, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9727626459143969, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9689922480620154, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04717560485005379, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9351969504447268, "eval_overall_precision": 0.9316455696202531, "eval_overall_recall": 0.9387755102040817, "eval_runtime": 0.2868, "eval_samples_per_second": 592.669, "eval_steps_per_second": 10.459, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.08714733272790909, "learning_rate": 3.7e-05, "loss": 0.031, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9171270718232045, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8736842105263158, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9230769230769231, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9364161849710982, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04282020777463913, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2852, "eval_samples_per_second": 596.048, "eval_steps_per_second": 10.518, "step": 2496 }, { "epoch": 27.0, "grad_norm": 0.43710198998451233, "learning_rate": 3.65e-05, "loss": 0.0302, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.8983957219251336, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8316831683168316, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9147727272727272, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9252873563218391, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.051920317113399506, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9308176100628932, "eval_overall_precision": 0.9181141439205955, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2845, "eval_samples_per_second": 597.548, "eval_steps_per_second": 10.545, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.4664277732372284, "learning_rate": 3.6e-05, "loss": 0.0288, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9152542372881357, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8901098901098901, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9411764705882353, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9385474860335196, "eval_ORGANIZATION_recall": 0.9438202247191011, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04365524277091026, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9480354879594423, "eval_overall_precision": 0.9420654911838791, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.2877, "eval_samples_per_second": 590.884, "eval_steps_per_second": 10.427, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.24498505890369415, "learning_rate": 3.55e-05, "loss": 0.0282, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9265536723163842, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9010989010989011, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9441340782122906, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9388888888888889, "eval_ORGANIZATION_recall": 0.949438202247191, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04060957208275795, "eval_overall_accuracy": 0.9902834008097166, "eval_overall_f1": 0.9518987341772152, "eval_overall_precision": 0.9447236180904522, "eval_overall_recall": 0.9591836734693877, "eval_runtime": 0.2868, "eval_samples_per_second": 592.749, "eval_steps_per_second": 10.46, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.4071507155895233, "learning_rate": 3.5e-05, "loss": 0.0279, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.9021739130434784, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8469387755102041, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9485714285714285, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9651162790697675, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05195392295718193, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.9468354430379746, "eval_overall_precision": 0.9396984924623115, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.2868, "eval_samples_per_second": 592.644, "eval_steps_per_second": 10.458, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.060660116374492645, "learning_rate": 3.45e-05, "loss": 0.0261, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.8983957219251336, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8316831683168316, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9173789173789175, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.930635838150289, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04885789752006531, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9319899244332494, "eval_overall_precision": 0.9203980099502488, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.289, "eval_samples_per_second": 588.215, "eval_steps_per_second": 10.38, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.9730001091957092, "learning_rate": 3.4000000000000007e-05, "loss": 0.0239, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9010989010989011, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8541666666666666, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9235127478753541, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9314285714285714, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.0425848551094532, "eval_overall_accuracy": 0.9889338731443995, "eval_overall_f1": 0.9355246523388117, "eval_overall_precision": 0.9273182957393483, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2918, "eval_samples_per_second": 582.529, "eval_steps_per_second": 10.28, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.8699120879173279, "learning_rate": 3.35e-05, "loss": 0.0262, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.8999999999999999, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8617021276595744, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9265536723163842, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9318181818181818, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05084100738167763, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9367088607594937, "eval_overall_precision": 0.9296482412060302, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2871, "eval_samples_per_second": 592.208, "eval_steps_per_second": 10.451, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.08233151584863663, "learning_rate": 3.3e-05, "loss": 0.0233, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.9050279329608939, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8709677419354839, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9291784702549575, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9371428571428572, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05224703624844551, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2869, "eval_samples_per_second": 592.499, "eval_steps_per_second": 10.456, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.5700938701629639, "learning_rate": 3.2500000000000004e-05, "loss": 0.0234, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.907103825136612, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8556701030927835, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.942857142857143, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9593023255813954, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04535231739282608, "eval_overall_accuracy": 0.9889338731443995, "eval_overall_f1": 0.9467005076142132, "eval_overall_precision": 0.9419191919191919, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.286, "eval_samples_per_second": 594.352, "eval_steps_per_second": 10.489, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.6118347644805908, "learning_rate": 3.2000000000000005e-05, "loss": 0.0242, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9575070821529745, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9657142857142857, "eval_ORGANIZATION_recall": 0.949438202247191, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04519454017281532, "eval_overall_accuracy": 0.9892037786774629, "eval_overall_f1": 0.9531051964512041, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9591836734693877, "eval_runtime": 0.2873, "eval_samples_per_second": 591.669, "eval_steps_per_second": 10.441, "step": 3456 }, { "epoch": 37.0, "grad_norm": 0.45790621638298035, "learning_rate": 3.15e-05, "loss": 0.0223, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.9111111111111112, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8723404255319149, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9378531073446328, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9431818181818182, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04222346469759941, "eval_overall_accuracy": 0.9892037786774629, "eval_overall_f1": 0.9443037974683544, "eval_overall_precision": 0.9371859296482412, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2862, "eval_samples_per_second": 593.973, "eval_steps_per_second": 10.482, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.19207726418972015, "learning_rate": 3.1e-05, "loss": 0.0225, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.8817204301075269, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.82, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9169054441260744, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.935672514619883, "eval_ORGANIZATION_recall": 0.898876404494382, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.048989761620759964, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.9279393173198484, "eval_overall_precision": 0.9197994987468672, "eval_overall_recall": 0.9362244897959183, "eval_runtime": 0.2875, "eval_samples_per_second": 591.343, "eval_steps_per_second": 10.435, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.19408129155635834, "learning_rate": 3.05e-05, "loss": 0.0233, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.908108108108108, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8484848484848485, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9325842696629213, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9325842696629213, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05212869122624397, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9410288582183187, "eval_overall_precision": 0.9259259259259259, "eval_overall_recall": 0.9566326530612245, "eval_runtime": 0.2889, "eval_samples_per_second": 588.505, "eval_steps_per_second": 10.385, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.1511041224002838, "learning_rate": 3e-05, "loss": 0.0197, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.9050279329608939, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8709677419354839, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9362880886426592, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9234972677595629, "eval_ORGANIZATION_recall": 0.949438202247191, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.048837099224328995, "eval_overall_accuracy": 0.9883940620782726, "eval_overall_f1": 0.9433962264150944, "eval_overall_precision": 0.9305210918114144, "eval_overall_recall": 0.9566326530612245, "eval_runtime": 0.2873, "eval_samples_per_second": 591.774, "eval_steps_per_second": 10.443, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.0778215155005455, "learning_rate": 2.95e-05, "loss": 0.0205, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9352112676056338, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9378531073446328, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04943624511361122, "eval_overall_accuracy": 0.9883940620782726, "eval_overall_f1": 0.9443037974683544, "eval_overall_precision": 0.9371859296482412, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2874, "eval_samples_per_second": 591.444, "eval_steps_per_second": 10.437, "step": 3936 }, { "epoch": 42.0, "grad_norm": 1.0895295143127441, "learning_rate": 2.9e-05, "loss": 0.0206, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.9142857142857143, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.898876404494382, "eval_LOCATION_recall": 0.9302325581395349, "eval_ORGANIZATION_f1": 0.9352112676056338, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9378531073446328, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04356543719768524, "eval_overall_accuracy": 0.988663967611336, "eval_overall_f1": 0.9440203562340966, "eval_overall_precision": 0.9416243654822335, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2866, "eval_samples_per_second": 593.189, "eval_steps_per_second": 10.468, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.10658109933137894, "learning_rate": 2.8499999999999998e-05, "loss": 0.0194, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9111111111111112, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8723404255319149, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9405099150141643, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9485714285714286, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.047641169279813766, "eval_overall_accuracy": 0.9889338731443995, "eval_overall_f1": 0.9455006337135614, "eval_overall_precision": 0.9395465994962217, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2889, "eval_samples_per_second": 588.521, "eval_steps_per_second": 10.386, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.7513548731803894, "learning_rate": 2.8000000000000003e-05, "loss": 0.019, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.8961748633879781, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.845360824742268, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9310344827586207, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9529411764705882, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05172840505838394, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9377382465057178, "eval_overall_precision": 0.9341772151898734, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2886, "eval_samples_per_second": 589.07, "eval_steps_per_second": 10.395, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.8231066465377808, "learning_rate": 2.7500000000000004e-05, "loss": 0.0199, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9431818181818182, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9540229885057471, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.049038875848054886, "eval_overall_accuracy": 0.9881241565452091, "eval_overall_f1": 0.9455006337135614, "eval_overall_precision": 0.9395465994962217, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2858, "eval_samples_per_second": 594.841, "eval_steps_per_second": 10.497, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.19870945811271667, "learning_rate": 2.7000000000000002e-05, "loss": 0.0165, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.9180327868852459, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.865979381443299, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9226361031518625, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9415204678362573, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.057863056659698486, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2884, "eval_samples_per_second": 589.486, "eval_steps_per_second": 10.403, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.2438984364271164, "learning_rate": 2.6500000000000004e-05, "loss": 0.0178, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.9010989010989011, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8541666666666666, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9287749287749287, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9421965317919075, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.052241381257772446, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.285, "eval_samples_per_second": 596.467, "eval_steps_per_second": 10.526, "step": 4512 }, { "epoch": 48.0, "grad_norm": 2.206496238708496, "learning_rate": 2.6000000000000002e-05, "loss": 0.0173, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9352112676056338, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9378531073446328, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05471523106098175, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.943109987357775, "eval_overall_precision": 0.9348370927318296, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2856, "eval_samples_per_second": 595.237, "eval_steps_per_second": 10.504, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.10019052773714066, "learning_rate": 2.5500000000000003e-05, "loss": 0.0176, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.942857142857143, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9593023255813954, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.0536094605922699, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9453621346886911, "eval_overall_precision": 0.9417721518987342, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2876, "eval_samples_per_second": 591.131, "eval_steps_per_second": 10.432, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.3046879768371582, "learning_rate": 2.5e-05, "loss": 0.0173, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.8926553672316384, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8681318681318682, "eval_LOCATION_recall": 0.9186046511627907, "eval_ORGANIZATION_f1": 0.9405099150141643, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9485714285714286, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05412731319665909, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9426751592356687, "eval_overall_precision": 0.9414758269720102, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2877, "eval_samples_per_second": 590.925, "eval_steps_per_second": 10.428, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.5300818681716919, "learning_rate": 2.45e-05, "loss": 0.016, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.893854748603352, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8602150537634409, "eval_LOCATION_recall": 0.9302325581395349, "eval_ORGANIZATION_f1": 0.9164265129682998, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9408284023668639, "eval_ORGANIZATION_recall": 0.8932584269662921, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.053407933562994, "eval_overall_accuracy": 0.9851551956815114, "eval_overall_f1": 0.9309462915601023, "eval_overall_precision": 0.9333333333333333, "eval_overall_recall": 0.9285714285714286, "eval_runtime": 0.2844, "eval_samples_per_second": 597.677, "eval_steps_per_second": 10.547, "step": 4896 }, { "epoch": 52.0, "grad_norm": 0.40124499797821045, "learning_rate": 2.4e-05, "loss": 0.0144, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9050279329608939, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8709677419354839, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.056055791676044464, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9389312977099236, "eval_overall_precision": 0.9365482233502538, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2853, "eval_samples_per_second": 595.839, "eval_steps_per_second": 10.515, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.657749593257904, "learning_rate": 2.35e-05, "loss": 0.0163, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9171270718232045, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8736842105263158, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9295774647887324, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9322033898305084, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06042208895087242, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9419191919191919, "eval_overall_precision": 0.9325, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2853, "eval_samples_per_second": 595.767, "eval_steps_per_second": 10.514, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.1539894938468933, "learning_rate": 2.3000000000000003e-05, "loss": 0.0157, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.9171270718232045, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8736842105263158, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05426372215151787, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9404309252217997, "eval_overall_precision": 0.9345088161209067, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2861, "eval_samples_per_second": 594.212, "eval_steps_per_second": 10.486, "step": 5184 }, { "epoch": 55.0, "grad_norm": 0.32259437441825867, "learning_rate": 2.25e-05, "loss": 0.015, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.875, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9261363636363636, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9367816091954023, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05390578880906105, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9417721518987343, "eval_overall_precision": 0.9346733668341709, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2856, "eval_samples_per_second": 595.161, "eval_steps_per_second": 10.503, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.5262906551361084, "learning_rate": 2.2000000000000003e-05, "loss": 0.0124, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9199999999999999, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.936046511627907, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05794522911310196, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9365482233502538, "eval_overall_precision": 0.9318181818181818, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2859, "eval_samples_per_second": 594.585, "eval_steps_per_second": 10.493, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.7698050737380981, "learning_rate": 2.15e-05, "loss": 0.0143, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.9032258064516129, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.84, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9096045197740114, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9147727272727273, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05896604061126709, "eval_overall_accuracy": 0.9859649122807017, "eval_overall_f1": 0.9296482412060302, "eval_overall_precision": 0.9158415841584159, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2874, "eval_samples_per_second": 591.525, "eval_steps_per_second": 10.439, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.1310461163520813, "learning_rate": 2.1e-05, "loss": 0.014, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9180327868852459, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.865979381443299, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9226361031518625, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9415204678362573, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06076360121369362, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2854, "eval_samples_per_second": 595.564, "eval_steps_per_second": 10.51, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.25326842069625854, "learning_rate": 2.05e-05, "loss": 0.0139, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9111111111111112, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8723404255319149, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9408450704225352, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.943502824858757, "eval_ORGANIZATION_recall": 0.9382022471910112, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.055396970361471176, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9456384323640962, "eval_overall_precision": 0.9373433583959899, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.2846, "eval_samples_per_second": 597.225, "eval_steps_per_second": 10.539, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.4024540185928345, "learning_rate": 2e-05, "loss": 0.0147, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.907103825136612, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8556701030927835, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9344729344729344, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9479768786127167, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05850914120674133, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9429657794676806, "eval_overall_precision": 0.9370277078085643, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2844, "eval_samples_per_second": 597.794, "eval_steps_per_second": 10.549, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.25083741545677185, "learning_rate": 1.9500000000000003e-05, "loss": 0.0136, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.9111111111111112, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8723404255319149, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9408450704225352, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.943502824858757, "eval_ORGANIZATION_recall": 0.9382022471910112, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05618133023381233, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9456384323640962, "eval_overall_precision": 0.9373433583959899, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.2844, "eval_samples_per_second": 597.741, "eval_steps_per_second": 10.548, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.43566325306892395, "learning_rate": 1.9e-05, "loss": 0.0134, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.9222222222222223, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8829787234042553, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.935933147632312, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9281767955801105, "eval_ORGANIZATION_recall": 0.9438202247191011, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05263877287507057, "eval_overall_accuracy": 0.9883940620782726, "eval_overall_f1": 0.9459119496855347, "eval_overall_precision": 0.9330024813895782, "eval_overall_recall": 0.9591836734693877, "eval_runtime": 0.2854, "eval_samples_per_second": 595.585, "eval_steps_per_second": 10.51, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.9059112071990967, "learning_rate": 1.85e-05, "loss": 0.014, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.9273743016759777, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8924731182795699, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9329608938547486, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9277777777777778, "eval_ORGANIZATION_recall": 0.9382022471910112, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05600650608539581, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9457755359394704, "eval_overall_precision": 0.9351620947630923, "eval_overall_recall": 0.9566326530612245, "eval_runtime": 0.2855, "eval_samples_per_second": 595.41, "eval_steps_per_second": 10.507, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.3811323046684265, "learning_rate": 1.8e-05, "loss": 0.0136, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.935933147632312, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9281767955801105, "eval_ORGANIZATION_recall": 0.9438202247191011, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.04886498302221298, "eval_overall_accuracy": 0.9892037786774629, "eval_overall_f1": 0.9457755359394704, "eval_overall_precision": 0.9351620947630923, "eval_overall_recall": 0.9566326530612245, "eval_runtime": 0.2867, "eval_samples_per_second": 592.984, "eval_steps_per_second": 10.464, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.1959855705499649, "learning_rate": 1.75e-05, "loss": 0.0124, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.9273743016759777, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8924731182795699, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.96045197740113, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9659090909090909, "eval_ORGANIZATION_recall": 0.9550561797752809, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05504164472222328, "eval_overall_accuracy": 0.9894736842105263, "eval_overall_f1": 0.9593908629441624, "eval_overall_precision": 0.9545454545454546, "eval_overall_recall": 0.9642857142857143, "eval_runtime": 0.2849, "eval_samples_per_second": 596.763, "eval_steps_per_second": 10.531, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.23753449320793152, "learning_rate": 1.7000000000000003e-05, "loss": 0.0117, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9431818181818182, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9540229885057471, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05207521840929985, "eval_overall_accuracy": 0.988663967611336, "eval_overall_f1": 0.9455006337135614, "eval_overall_precision": 0.9395465994962217, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.286, "eval_samples_per_second": 594.42, "eval_steps_per_second": 10.49, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.026631498709321022, "learning_rate": 1.65e-05, "loss": 0.0127, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9545454545454545, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9655172413793104, "eval_ORGANIZATION_recall": 0.9438202247191011, "eval_PERSON_f1": 0.9803921568627452, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.984251968503937, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05086338892579079, "eval_overall_accuracy": 0.9892037786774629, "eval_overall_f1": 0.9517766497461929, "eval_overall_precision": 0.946969696969697, "eval_overall_recall": 0.9566326530612245, "eval_runtime": 0.2845, "eval_samples_per_second": 597.452, "eval_steps_per_second": 10.543, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.05703957751393318, "learning_rate": 1.6000000000000003e-05, "loss": 0.012, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9204545454545455, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9418604651162791, "eval_ORGANIZATION_f1": 0.935933147632312, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9281767955801105, "eval_ORGANIZATION_recall": 0.9438202247191011, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05626333877444267, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9456384323640962, "eval_overall_precision": 0.9373433583959899, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.2859, "eval_samples_per_second": 594.629, "eval_steps_per_second": 10.493, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.6835038661956787, "learning_rate": 1.55e-05, "loss": 0.0118, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.9180327868852459, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.865979381443299, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9375000000000001, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9482758620689655, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.054845135658979416, "eval_overall_accuracy": 0.988663967611336, "eval_overall_f1": 0.9456384323640962, "eval_overall_precision": 0.9373433583959899, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.2853, "eval_samples_per_second": 595.873, "eval_steps_per_second": 10.515, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.3324280381202698, "learning_rate": 1.5e-05, "loss": 0.0119, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.907103825136612, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8556701030927835, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9322033898305084, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9375, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.059074658900499344, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9407313997477933, "eval_overall_precision": 0.9301745635910225, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2853, "eval_samples_per_second": 595.849, "eval_steps_per_second": 10.515, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.41472673416137695, "learning_rate": 1.45e-05, "loss": 0.0119, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9213483146067417, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8913043478260869, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9388888888888889, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9285714285714286, "eval_ORGANIZATION_recall": 0.949438202247191, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.060846418142318726, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.947103274559194, "eval_overall_precision": 0.9353233830845771, "eval_overall_recall": 0.9591836734693877, "eval_runtime": 0.2857, "eval_samples_per_second": 594.945, "eval_steps_per_second": 10.499, "step": 6816 }, { "epoch": 72.0, "grad_norm": 1.2578948736190796, "learning_rate": 1.4000000000000001e-05, "loss": 0.0102, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.9162011173184358, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8817204301075269, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9382022471910112, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9382022471910112, "eval_ORGANIZATION_recall": 0.9382022471910112, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.056985922157764435, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9456384323640962, "eval_overall_precision": 0.9373433583959899, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.2874, "eval_samples_per_second": 591.423, "eval_steps_per_second": 10.437, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.3907749354839325, "learning_rate": 1.3500000000000001e-05, "loss": 0.0108, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9130434782608695, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8571428571428571, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9287749287749287, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9421965317919075, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06016295775771141, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9405815423514539, "eval_overall_precision": 0.9323308270676691, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2875, "eval_samples_per_second": 591.247, "eval_steps_per_second": 10.434, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.14214874804019928, "learning_rate": 1.3000000000000001e-05, "loss": 0.0112, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9010989010989011, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8541666666666666, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9314285714285714, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9476744186046512, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06054116040468216, "eval_overall_accuracy": 0.9870445344129555, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2859, "eval_samples_per_second": 594.654, "eval_steps_per_second": 10.494, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.3518529236316681, "learning_rate": 1.25e-05, "loss": 0.011, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.9032258064516129, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.84, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9002849002849002, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9132947976878613, "eval_ORGANIZATION_recall": 0.8876404494382022, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06660357862710953, "eval_overall_accuracy": 0.9856950067476383, "eval_overall_f1": 0.9255989911727616, "eval_overall_precision": 0.9152119700748129, "eval_overall_recall": 0.9362244897959183, "eval_runtime": 0.291, "eval_samples_per_second": 584.169, "eval_steps_per_second": 10.309, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.3386372923851013, "learning_rate": 1.2e-05, "loss": 0.0098, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9111111111111112, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8723404255319149, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9325842696629213, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9325842696629213, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.057027120143175125, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9419191919191919, "eval_overall_precision": 0.9325, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2888, "eval_samples_per_second": 588.564, "eval_steps_per_second": 10.386, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.010590732097625732, "learning_rate": 1.1500000000000002e-05, "loss": 0.0098, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9295774647887324, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9322033898305084, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05786881595849991, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9407313997477933, "eval_overall_precision": 0.9301745635910225, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2865, "eval_samples_per_second": 593.392, "eval_steps_per_second": 10.472, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.1329515427350998, "learning_rate": 1.1000000000000001e-05, "loss": 0.0099, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9130434782608695, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8571428571428571, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9235127478753541, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9314285714285714, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06257299333810806, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9382093316519546, "eval_overall_precision": 0.9276807980049875, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2878, "eval_samples_per_second": 590.613, "eval_steps_per_second": 10.423, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.14144736528396606, "learning_rate": 1.05e-05, "loss": 0.0107, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9222222222222223, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8829787234042553, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9235127478753541, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9314285714285714, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.058039430528879166, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9404309252217997, "eval_overall_precision": 0.9345088161209067, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2865, "eval_samples_per_second": 593.452, "eval_steps_per_second": 10.473, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.27334415912628174, "learning_rate": 1e-05, "loss": 0.0103, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9180327868852459, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.865979381443299, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9325842696629213, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9325842696629213, "eval_ORGANIZATION_recall": 0.9325842696629213, "eval_PERSON_f1": 0.96875, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.96875, "eval_PERSON_recall": 0.96875, "eval_loss": 0.05926325172185898, "eval_overall_accuracy": 0.9881241565452091, "eval_overall_f1": 0.9408805031446542, "eval_overall_precision": 0.9280397022332506, "eval_overall_recall": 0.9540816326530612, "eval_runtime": 0.2856, "eval_samples_per_second": 595.151, "eval_steps_per_second": 10.503, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.09895119816064835, "learning_rate": 9.5e-06, "loss": 0.0091, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.9180327868852459, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.865979381443299, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9461756373937678, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9542857142857143, "eval_ORGANIZATION_recall": 0.9382022471910112, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06263738870620728, "eval_overall_accuracy": 0.988663967611336, "eval_overall_f1": 0.9494949494949495, "eval_overall_precision": 0.94, "eval_overall_recall": 0.9591836734693877, "eval_runtime": 0.2881, "eval_samples_per_second": 590.102, "eval_steps_per_second": 10.414, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.2618274986743927, "learning_rate": 9e-06, "loss": 0.0105, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9235127478753541, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9314285714285714, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05788002163171768, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.9380530973451328, "eval_overall_precision": 0.9298245614035088, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2887, "eval_samples_per_second": 588.941, "eval_steps_per_second": 10.393, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.7836906909942627, "learning_rate": 8.500000000000002e-06, "loss": 0.0103, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9180327868852459, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.865979381443299, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9279538904899135, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9526627218934911, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06573164463043213, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.94147582697201, "eval_overall_precision": 0.9390862944162437, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2889, "eval_samples_per_second": 588.437, "eval_steps_per_second": 10.384, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.07423603534698486, "learning_rate": 8.000000000000001e-06, "loss": 0.0118, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9180327868852459, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.865979381443299, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9230769230769231, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9364161849710982, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06355427950620651, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9392405063291138, "eval_overall_precision": 0.9321608040201005, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2858, "eval_samples_per_second": 594.862, "eval_steps_per_second": 10.498, "step": 8064 }, { "epoch": 85.0, "grad_norm": 1.2200062274932861, "learning_rate": 7.5e-06, "loss": 0.0095, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.9060773480662985, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8631578947368421, "eval_LOCATION_recall": 0.9534883720930233, "eval_ORGANIZATION_f1": 0.9204545454545454, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9310344827586207, "eval_ORGANIZATION_recall": 0.9101123595505618, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06323674321174622, "eval_overall_accuracy": 0.9867746288798921, "eval_overall_f1": 0.935361216730038, "eval_overall_precision": 0.929471032745592, "eval_overall_recall": 0.9413265306122449, "eval_runtime": 0.2906, "eval_samples_per_second": 585.065, "eval_steps_per_second": 10.325, "step": 8160 }, { "epoch": 86.0, "grad_norm": 0.8143560886383057, "learning_rate": 7.000000000000001e-06, "loss": 0.0089, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9180327868852459, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.865979381443299, "eval_LOCATION_recall": 0.9767441860465116, "eval_ORGANIZATION_f1": 0.9226361031518625, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9415204678362573, "eval_ORGANIZATION_recall": 0.9044943820224719, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06056205928325653, "eval_overall_accuracy": 0.9862348178137652, "eval_overall_f1": 0.9390862944162437, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9438775510204082, "eval_runtime": 0.2913, "eval_samples_per_second": 583.568, "eval_steps_per_second": 10.298, "step": 8256 }, { "epoch": 87.0, "grad_norm": 1.2203452587127686, "learning_rate": 6.5000000000000004e-06, "loss": 0.0086, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.907103825136612, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8556701030927835, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9344729344729344, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9479768786127167, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.059925343841314316, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9417721518987343, "eval_overall_precision": 0.9346733668341709, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2906, "eval_samples_per_second": 584.911, "eval_steps_per_second": 10.322, "step": 8352 }, { "epoch": 88.0, "grad_norm": 1.5901998281478882, "learning_rate": 6e-06, "loss": 0.0104, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9375000000000001, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9482758620689655, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.059619657695293427, "eval_overall_accuracy": 0.9892037786774629, "eval_overall_f1": 0.9443037974683544, "eval_overall_precision": 0.9371859296482412, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2972, "eval_samples_per_second": 571.995, "eval_steps_per_second": 10.094, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.22172300517559052, "learning_rate": 5.500000000000001e-06, "loss": 0.0081, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9375000000000001, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9482758620689655, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05831991136074066, "eval_overall_accuracy": 0.9894736842105263, "eval_overall_f1": 0.9443037974683544, "eval_overall_precision": 0.9371859296482412, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2907, "eval_samples_per_second": 584.758, "eval_steps_per_second": 10.319, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.11473022401332855, "learning_rate": 5e-06, "loss": 0.0076, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.934844192634561, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9428571428571428, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05891711264848709, "eval_overall_accuracy": 0.988663967611336, "eval_overall_f1": 0.943109987357775, "eval_overall_precision": 0.9348370927318296, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2905, "eval_samples_per_second": 585.276, "eval_steps_per_second": 10.328, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.028633061796426773, "learning_rate": 4.5e-06, "loss": 0.0092, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9318181818181818, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9425287356321839, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.061169691383838654, "eval_overall_accuracy": 0.9881241565452091, "eval_overall_f1": 0.9417721518987343, "eval_overall_precision": 0.9346733668341709, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2898, "eval_samples_per_second": 586.56, "eval_steps_per_second": 10.351, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.1511673629283905, "learning_rate": 4.000000000000001e-06, "loss": 0.008, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9344729344729344, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9479768786127167, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05958572402596474, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9429657794676806, "eval_overall_precision": 0.9370277078085643, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2936, "eval_samples_per_second": 578.922, "eval_steps_per_second": 10.216, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.7097914814949036, "learning_rate": 3.5000000000000004e-06, "loss": 0.0086, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9235127478753541, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9314285714285714, "eval_ORGANIZATION_recall": 0.9157303370786517, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.060184404253959656, "eval_overall_accuracy": 0.9865047233468286, "eval_overall_f1": 0.9380530973451328, "eval_overall_precision": 0.9298245614035088, "eval_overall_recall": 0.9464285714285714, "eval_runtime": 0.2899, "eval_samples_per_second": 586.429, "eval_steps_per_second": 10.349, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.06405020505189896, "learning_rate": 3e-06, "loss": 0.0063, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.942857142857143, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9593023255813954, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06053685396909714, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9467005076142132, "eval_overall_precision": 0.9419191919191919, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2915, "eval_samples_per_second": 583.255, "eval_steps_per_second": 10.293, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.5586755275726318, "learning_rate": 2.5e-06, "loss": 0.0078, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.9171270718232045, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8736842105263158, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9461756373937678, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9542857142857143, "eval_ORGANIZATION_recall": 0.9382022471910112, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.059372350573539734, "eval_overall_accuracy": 0.988663967611336, "eval_overall_f1": 0.949367088607595, "eval_overall_precision": 0.9422110552763819, "eval_overall_recall": 0.9566326530612245, "eval_runtime": 0.2893, "eval_samples_per_second": 587.53, "eval_steps_per_second": 10.368, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.006612082477658987, "learning_rate": 2.0000000000000003e-06, "loss": 0.0079, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.9171270718232045, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8736842105263158, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9375000000000001, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9482758620689655, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05879458412528038, "eval_overall_accuracy": 0.9881241565452091, "eval_overall_f1": 0.9455006337135614, "eval_overall_precision": 0.9395465994962217, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2908, "eval_samples_per_second": 584.627, "eval_steps_per_second": 10.317, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.007970322854816914, "learning_rate": 1.5e-06, "loss": 0.0083, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.942857142857143, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9593023255813954, "eval_ORGANIZATION_recall": 0.9269662921348315, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06146546080708504, "eval_overall_accuracy": 0.9881241565452091, "eval_overall_f1": 0.9467005076142132, "eval_overall_precision": 0.9419191919191919, "eval_overall_recall": 0.951530612244898, "eval_runtime": 0.2891, "eval_samples_per_second": 587.978, "eval_steps_per_second": 10.376, "step": 9312 }, { "epoch": 98.0, "grad_norm": 0.5247291326522827, "learning_rate": 1.0000000000000002e-06, "loss": 0.0075, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9371428571428573, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9534883720930233, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06113233417272568, "eval_overall_accuracy": 0.9873144399460189, "eval_overall_f1": 0.9441624365482234, "eval_overall_precision": 0.9393939393939394, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2903, "eval_samples_per_second": 585.602, "eval_steps_per_second": 10.334, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.746631920337677, "learning_rate": 5.000000000000001e-07, "loss": 0.0083, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9371428571428573, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9534883720930233, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.06001231074333191, "eval_overall_accuracy": 0.9875843454790824, "eval_overall_f1": 0.9441624365482234, "eval_overall_precision": 0.9393939393939394, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2904, "eval_samples_per_second": 585.452, "eval_steps_per_second": 10.332, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.05131411924958229, "learning_rate": 0.0, "loss": 0.0074, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9120879120879122, "eval_LOCATION_number": 86, "eval_LOCATION_precision": 0.8645833333333334, "eval_LOCATION_recall": 0.9651162790697675, "eval_ORGANIZATION_f1": 0.9344729344729344, "eval_ORGANIZATION_number": 178, "eval_ORGANIZATION_precision": 0.9479768786127167, "eval_ORGANIZATION_recall": 0.9213483146067416, "eval_PERSON_f1": 0.9765625, "eval_PERSON_number": 128, "eval_PERSON_precision": 0.9765625, "eval_PERSON_recall": 0.9765625, "eval_loss": 0.05995160713791847, "eval_overall_accuracy": 0.9878542510121457, "eval_overall_f1": 0.9429657794676806, "eval_overall_precision": 0.9370277078085643, "eval_overall_recall": 0.9489795918367347, "eval_runtime": 0.2939, "eval_samples_per_second": 578.339, "eval_steps_per_second": 10.206, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 4315798421360676.0, "train_loss": 0.038009579268594586, "train_runtime": 536.2277, "train_samples_per_second": 285.513, "train_steps_per_second": 17.903 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 4315798421360676.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }