{ "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": 1.4604038000106812, "learning_rate": 4.9500000000000004e-05, "loss": 1.0507, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.0, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.0, "eval_LOCATION_recall": 0.0, "eval_ORGANIZATION_f1": 0.0, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.0, "eval_ORGANIZATION_recall": 0.0, "eval_PERSON_f1": 0.0, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.0, "eval_PERSON_recall": 0.0, "eval_loss": 0.653229832649231, "eval_overall_accuracy": 0.8342935528120713, "eval_overall_f1": 0.0, "eval_overall_precision": 0.0, "eval_overall_recall": 0.0, "eval_runtime": 0.2536, "eval_samples_per_second": 670.386, "eval_steps_per_second": 11.83, "step": 96 }, { "epoch": 2.0, "grad_norm": 1.139877200126648, "learning_rate": 4.9e-05, "loss": 0.5943, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.0, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.0, "eval_LOCATION_recall": 0.0, "eval_ORGANIZATION_f1": 0.0748663101604278, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.3333333333333333, "eval_ORGANIZATION_recall": 0.04216867469879518, "eval_PERSON_f1": 0.15135135135135136, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.32558139534883723, "eval_PERSON_recall": 0.09859154929577464, "eval_loss": 0.4512753188610077, "eval_overall_accuracy": 0.842798353909465, "eval_overall_f1": 0.08993576017130621, "eval_overall_precision": 0.3181818181818182, "eval_overall_recall": 0.05236907730673317, "eval_runtime": 0.2451, "eval_samples_per_second": 693.551, "eval_steps_per_second": 12.239, "step": 192 }, { "epoch": 3.0, "grad_norm": 0.7932935953140259, "learning_rate": 4.85e-05, "loss": 0.4265, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.18840579710144925, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.28888888888888886, "eval_LOCATION_recall": 0.13978494623655913, "eval_ORGANIZATION_f1": 0.40431266846361186, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.36585365853658536, "eval_ORGANIZATION_recall": 0.45180722891566266, "eval_PERSON_f1": 0.4189944134078212, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.3472222222222222, "eval_PERSON_recall": 0.528169014084507, "eval_loss": 0.32607579231262207, "eval_overall_accuracy": 0.8954732510288066, "eval_overall_f1": 0.37600922722029995, "eval_overall_precision": 0.3497854077253219, "eval_overall_recall": 0.40648379052369077, "eval_runtime": 0.2475, "eval_samples_per_second": 686.819, "eval_steps_per_second": 12.12, "step": 288 }, { "epoch": 4.0, "grad_norm": 1.5371474027633667, "learning_rate": 4.8e-05, "loss": 0.3456, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.3313609467455621, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.3684210526315789, "eval_LOCATION_recall": 0.3010752688172043, "eval_ORGANIZATION_f1": 0.4781491002570694, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.4170403587443946, "eval_ORGANIZATION_recall": 0.5602409638554217, "eval_PERSON_f1": 0.4751381215469613, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.39090909090909093, "eval_PERSON_recall": 0.6056338028169014, "eval_loss": 0.2800743281841278, "eval_overall_accuracy": 0.9155006858710563, "eval_overall_f1": 0.45, "eval_overall_precision": 0.3988439306358382, "eval_overall_recall": 0.516209476309227, "eval_runtime": 0.2472, "eval_samples_per_second": 687.59, "eval_steps_per_second": 12.134, "step": 384 }, { "epoch": 5.0, "grad_norm": 1.2825340032577515, "learning_rate": 4.75e-05, "loss": 0.3004, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.3977900552486188, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.4090909090909091, "eval_LOCATION_recall": 0.3870967741935484, "eval_ORGANIZATION_f1": 0.5778894472361809, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.4956896551724138, "eval_ORGANIZATION_recall": 0.6927710843373494, "eval_PERSON_f1": 0.5878787878787879, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.5159574468085106, "eval_PERSON_recall": 0.6830985915492958, "eval_loss": 0.239501953125, "eval_overall_accuracy": 0.9336076817558299, "eval_overall_f1": 0.5456545654565457, "eval_overall_precision": 0.4881889763779528, "eval_overall_recall": 0.6184538653366584, "eval_runtime": 0.2472, "eval_samples_per_second": 687.806, "eval_steps_per_second": 12.138, "step": 480 }, { "epoch": 6.0, "grad_norm": 1.2340002059936523, "learning_rate": 4.7e-05, "loss": 0.2668, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.48704663212435234, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.47, "eval_LOCATION_recall": 0.5053763440860215, "eval_ORGANIZATION_f1": 0.6404199475065616, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.5674418604651162, "eval_ORGANIZATION_recall": 0.7349397590361446, "eval_PERSON_f1": 0.7102803738317757, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.6368715083798883, "eval_PERSON_recall": 0.8028169014084507, "eval_loss": 0.20743004977703094, "eval_overall_accuracy": 0.945679012345679, "eval_overall_f1": 0.6324022346368716, "eval_overall_precision": 0.5728744939271255, "eval_overall_recall": 0.7057356608478803, "eval_runtime": 0.2478, "eval_samples_per_second": 685.92, "eval_steps_per_second": 12.104, "step": 576 }, { "epoch": 7.0, "grad_norm": 0.9397792220115662, "learning_rate": 4.6500000000000005e-05, "loss": 0.2298, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.5353535353535354, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.5047619047619047, "eval_LOCATION_recall": 0.5698924731182796, "eval_ORGANIZATION_f1": 0.6918918918918918, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.6274509803921569, "eval_ORGANIZATION_recall": 0.7710843373493976, "eval_PERSON_f1": 0.7652733118971062, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.7041420118343196, "eval_PERSON_recall": 0.8380281690140845, "eval_loss": 0.17204177379608154, "eval_overall_accuracy": 0.9561042524005487, "eval_overall_f1": 0.6825938566552902, "eval_overall_precision": 0.6276150627615062, "eval_overall_recall": 0.7481296758104738, "eval_runtime": 0.2469, "eval_samples_per_second": 688.663, "eval_steps_per_second": 12.153, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.6758012771606445, "learning_rate": 4.600000000000001e-05, "loss": 0.1899, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.6192893401015228, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.5865384615384616, "eval_LOCATION_recall": 0.6559139784946236, "eval_ORGANIZATION_f1": 0.7507002801120448, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.7015706806282722, "eval_ORGANIZATION_recall": 0.8072289156626506, "eval_PERSON_f1": 0.8686868686868686, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.832258064516129, "eval_PERSON_recall": 0.9084507042253521, "eval_loss": 0.1367180198431015, "eval_overall_accuracy": 0.9640603566529492, "eval_overall_f1": 0.7614571092831962, "eval_overall_precision": 0.72, "eval_overall_recall": 0.8079800498753117, "eval_runtime": 0.2486, "eval_samples_per_second": 683.8, "eval_steps_per_second": 12.067, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.7296782732009888, "learning_rate": 4.55e-05, "loss": 0.1641, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.7058823529411764, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.6486486486486487, "eval_LOCATION_recall": 0.7741935483870968, "eval_ORGANIZATION_f1": 0.8080229226361031, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.7704918032786885, "eval_ORGANIZATION_recall": 0.8493975903614458, "eval_PERSON_f1": 0.9016949152542374, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.869281045751634, "eval_PERSON_recall": 0.9366197183098591, "eval_loss": 0.11540879309177399, "eval_overall_accuracy": 0.970644718792867, "eval_overall_f1": 0.8160377358490566, "eval_overall_precision": 0.7740492170022372, "eval_overall_recall": 0.8628428927680798, "eval_runtime": 0.2468, "eval_samples_per_second": 688.923, "eval_steps_per_second": 12.157, "step": 864 }, { "epoch": 10.0, "grad_norm": 1.2114511728286743, "learning_rate": 4.5e-05, "loss": 0.1436, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.7835051546391754, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7524752475247525, "eval_LOCATION_recall": 0.8172043010752689, "eval_ORGANIZATION_f1": 0.8275862068965518, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.7912087912087912, "eval_ORGANIZATION_recall": 0.8674698795180723, "eval_PERSON_f1": 0.9347079037800687, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.912751677852349, "eval_PERSON_recall": 0.9577464788732394, "eval_loss": 0.09992080926895142, "eval_overall_accuracy": 0.9753086419753086, "eval_overall_f1": 0.8547418967587035, "eval_overall_precision": 0.8240740740740741, "eval_overall_recall": 0.8877805486284289, "eval_runtime": 0.2465, "eval_samples_per_second": 689.694, "eval_steps_per_second": 12.171, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.7425956130027771, "learning_rate": 4.4500000000000004e-05, "loss": 0.1282, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.8020304568527918, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7596153846153846, "eval_LOCATION_recall": 0.8494623655913979, "eval_ORGANIZATION_f1": 0.829971181556196, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.7955801104972375, "eval_ORGANIZATION_recall": 0.8674698795180723, "eval_PERSON_f1": 0.9347079037800687, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.912751677852349, "eval_PERSON_recall": 0.9577464788732394, "eval_loss": 0.08958197385072708, "eval_overall_accuracy": 0.9758573388203018, "eval_overall_f1": 0.859880239520958, "eval_overall_precision": 0.8271889400921659, "eval_overall_recall": 0.8952618453865336, "eval_runtime": 0.2474, "eval_samples_per_second": 687.016, "eval_steps_per_second": 12.124, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.8763998746871948, "learning_rate": 4.4000000000000006e-05, "loss": 0.1193, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.81, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7570093457943925, "eval_LOCATION_recall": 0.8709677419354839, "eval_ORGANIZATION_f1": 0.830945558739255, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.7923497267759563, "eval_ORGANIZATION_recall": 0.8734939759036144, "eval_PERSON_f1": 0.9448275862068966, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9256756756756757, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.08479801565408707, "eval_overall_accuracy": 0.9755829903978052, "eval_overall_f1": 0.865315852205006, "eval_overall_precision": 0.8287671232876712, "eval_overall_recall": 0.9052369077306733, "eval_runtime": 0.2412, "eval_samples_per_second": 704.759, "eval_steps_per_second": 12.437, "step": 1152 }, { "epoch": 13.0, "grad_norm": 0.7340523600578308, "learning_rate": 4.35e-05, "loss": 0.1109, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.826530612244898, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7864077669902912, "eval_LOCATION_recall": 0.8709677419354839, "eval_ORGANIZATION_f1": 0.8579710144927536, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8268156424581006, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.9480968858131488, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9319727891156463, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.07661643624305725, "eval_overall_accuracy": 0.9786008230452675, "eval_overall_f1": 0.8819277108433734, "eval_overall_precision": 0.8531468531468531, "eval_overall_recall": 0.912718204488778, "eval_runtime": 0.2433, "eval_samples_per_second": 698.827, "eval_steps_per_second": 12.332, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.6258025765419006, "learning_rate": 4.3e-05, "loss": 0.1039, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.8426395939086294, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7980769230769231, "eval_LOCATION_recall": 0.8924731182795699, "eval_ORGANIZATION_f1": 0.8439306358381503, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8111111111111111, "eval_ORGANIZATION_recall": 0.8795180722891566, "eval_PERSON_f1": 0.9616724738675958, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9517241379310345, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.07304983586072922, "eval_overall_accuracy": 0.9786008230452675, "eval_overall_f1": 0.8843373493975903, "eval_overall_precision": 0.8554778554778555, "eval_overall_recall": 0.9152119700748129, "eval_runtime": 0.2398, "eval_samples_per_second": 708.948, "eval_steps_per_second": 12.511, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.7563566565513611, "learning_rate": 4.25e-05, "loss": 0.0974, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.8512820512820513, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8137254901960784, "eval_LOCATION_recall": 0.8924731182795699, "eval_ORGANIZATION_f1": 0.8505747126436781, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8131868131868132, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.9583333333333334, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9452054794520548, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.07149593532085419, "eval_overall_accuracy": 0.9775034293552812, "eval_overall_f1": 0.8880866425992779, "eval_overall_precision": 0.858139534883721, "eval_overall_recall": 0.9201995012468828, "eval_runtime": 0.2414, "eval_samples_per_second": 704.205, "eval_steps_per_second": 12.427, "step": 1440 }, { "epoch": 16.0, "grad_norm": 1.3999837636947632, "learning_rate": 4.2e-05, "loss": 0.0954, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.8383838383838383, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7904761904761904, "eval_LOCATION_recall": 0.8924731182795699, "eval_ORGANIZATION_f1": 0.853868194842407, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8142076502732241, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.9616724738675958, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9517241379310345, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.07120219618082047, "eval_overall_accuracy": 0.9777777777777777, "eval_overall_f1": 0.8872901678657075, "eval_overall_precision": 0.8545034642032333, "eval_overall_recall": 0.9226932668329177, "eval_runtime": 0.242, "eval_samples_per_second": 702.395, "eval_steps_per_second": 12.395, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.5972534418106079, "learning_rate": 4.15e-05, "loss": 0.0911, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.8717948717948718, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8333333333333334, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.8571428571428572, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8305084745762712, "eval_ORGANIZATION_recall": 0.8855421686746988, "eval_PERSON_f1": 0.9583333333333334, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9452054794520548, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06439894437789917, "eval_overall_accuracy": 0.9799725651577503, "eval_overall_f1": 0.8958837772397095, "eval_overall_precision": 0.8705882352941177, "eval_overall_recall": 0.9226932668329177, "eval_runtime": 0.2408, "eval_samples_per_second": 705.855, "eval_steps_per_second": 12.456, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.6317753195762634, "learning_rate": 4.1e-05, "loss": 0.0868, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.8704663212435233, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.84, "eval_LOCATION_recall": 0.9032258064516129, "eval_ORGANIZATION_f1": 0.8746355685131194, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.847457627118644, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06092643737792969, "eval_overall_accuracy": 0.9816186556927298, "eval_overall_f1": 0.9051094890510949, "eval_overall_precision": 0.8836104513064132, "eval_overall_recall": 0.9276807980049875, "eval_runtime": 0.2418, "eval_samples_per_second": 702.974, "eval_steps_per_second": 12.405, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.9092683792114258, "learning_rate": 4.05e-05, "loss": 0.082, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.8969072164948454, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8613861386138614, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8790560471976401, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.861271676300578, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.9583333333333334, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9452054794520548, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.059390630573034286, "eval_overall_accuracy": 0.9813443072702333, "eval_overall_f1": 0.9110840438489648, "eval_overall_precision": 0.8904761904761904, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2419, "eval_samples_per_second": 702.863, "eval_steps_per_second": 12.403, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.6492010951042175, "learning_rate": 4e-05, "loss": 0.0785, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.8911917098445595, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.86, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.8645533141210374, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8287292817679558, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05844908207654953, "eval_overall_accuracy": 0.9813443072702333, "eval_overall_f1": 0.9055690072639225, "eval_overall_precision": 0.88, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2478, "eval_samples_per_second": 686.08, "eval_steps_per_second": 12.107, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.6365026235580444, "learning_rate": 3.9500000000000005e-05, "loss": 0.0804, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.8923076923076922, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8529411764705882, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8695652173913043, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8379888268156425, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9583333333333334, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9452054794520548, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05869784578680992, "eval_overall_accuracy": 0.980246913580247, "eval_overall_f1": 0.9057971014492753, "eval_overall_precision": 0.8782201405152225, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.247, "eval_samples_per_second": 688.144, "eval_steps_per_second": 12.144, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.8496540784835815, "learning_rate": 3.9000000000000006e-05, "loss": 0.0751, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.9072164948453608, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8712871287128713, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.884393063583815, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.85, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9616724738675958, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9517241379310345, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05693579837679863, "eval_overall_accuracy": 0.9821673525377229, "eval_overall_f1": 0.9165659008464329, "eval_overall_precision": 0.8896713615023474, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.248, "eval_samples_per_second": 685.6, "eval_steps_per_second": 12.099, "step": 2112 }, { "epoch": 23.0, "grad_norm": 1.1801159381866455, "learning_rate": 3.85e-05, "loss": 0.0747, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.8934010152284263, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8461538461538461, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8942598187311178, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.896969696969697, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05449802055954933, "eval_overall_accuracy": 0.9829903978052126, "eval_overall_f1": 0.9189189189189189, "eval_overall_precision": 0.9055690072639225, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2476, "eval_samples_per_second": 686.678, "eval_steps_per_second": 12.118, "step": 2208 }, { "epoch": 24.0, "grad_norm": 0.9136117100715637, "learning_rate": 3.8e-05, "loss": 0.0733, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.8787878787878788, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8285714285714286, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8835820895522388, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8757396449704142, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.9583333333333334, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9452054794520548, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.055243976414203644, "eval_overall_accuracy": 0.9810699588477366, "eval_overall_f1": 0.9086479902557857, "eval_overall_precision": 0.888095238095238, "eval_overall_recall": 0.9301745635910225, "eval_runtime": 0.2464, "eval_samples_per_second": 690.056, "eval_steps_per_second": 12.177, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.6865504384040833, "learning_rate": 3.7500000000000003e-05, "loss": 0.0695, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.8888888888888888, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8380952380952381, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8680351906158359, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8457142857142858, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.9583333333333334, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9452054794520548, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.054658565670251846, "eval_overall_accuracy": 0.9816186556927298, "eval_overall_f1": 0.9044740024183797, "eval_overall_precision": 0.8779342723004695, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2468, "eval_samples_per_second": 688.777, "eval_steps_per_second": 12.155, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.20866434276103973, "learning_rate": 3.7e-05, "loss": 0.0697, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.8969072164948454, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8613861386138614, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8862275449101796, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8809523809523809, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.9616724738675958, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9517241379310345, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04981612041592598, "eval_overall_accuracy": 0.9838134430727024, "eval_overall_f1": 0.9153374233128836, "eval_overall_precision": 0.9009661835748792, "eval_overall_recall": 0.9301745635910225, "eval_runtime": 0.2469, "eval_samples_per_second": 688.502, "eval_steps_per_second": 12.15, "step": 2496 }, { "epoch": 27.0, "grad_norm": 0.8859153389930725, "learning_rate": 3.65e-05, "loss": 0.064, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9072164948453608, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8712871287128713, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8856304985337244, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8628571428571429, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9583333333333334, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9452054794520548, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.048702120780944824, "eval_overall_accuracy": 0.9838134430727024, "eval_overall_f1": 0.9161603888213851, "eval_overall_precision": 0.8933649289099526, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2414, "eval_samples_per_second": 704.354, "eval_steps_per_second": 12.43, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.5901989340782166, "learning_rate": 3.6e-05, "loss": 0.0663, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9015544041450777, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.87, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8800000000000001, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8369565217391305, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05010833591222763, "eval_overall_accuracy": 0.9832647462277092, "eval_overall_f1": 0.9143546441495778, "eval_overall_precision": 0.8855140186915887, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2414, "eval_samples_per_second": 704.089, "eval_steps_per_second": 12.425, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.45621567964553833, "learning_rate": 3.55e-05, "loss": 0.064, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.8979591836734693, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8543689320388349, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8849557522123894, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8670520231213873, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0482449010014534, "eval_overall_accuracy": 0.9843621399176955, "eval_overall_f1": 0.9159561510353227, "eval_overall_precision": 0.8952380952380953, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2421, "eval_samples_per_second": 702.234, "eval_steps_per_second": 12.392, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.7974165081977844, "learning_rate": 3.5e-05, "loss": 0.0606, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.9025641025641027, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8627450980392157, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8895348837209304, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8595505617977528, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.047746580094099045, "eval_overall_accuracy": 0.9840877914951989, "eval_overall_f1": 0.9187878787878788, "eval_overall_precision": 0.8938679245283019, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2419, "eval_samples_per_second": 702.876, "eval_steps_per_second": 12.404, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.22428223490715027, "learning_rate": 3.45e-05, "loss": 0.0616, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.911917098445596, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.88, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.880952380952381, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8705882352941177, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04754789173603058, "eval_overall_accuracy": 0.9854595336076818, "eval_overall_f1": 0.9177914110429448, "eval_overall_precision": 0.9033816425120773, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.24, "eval_samples_per_second": 708.439, "eval_steps_per_second": 12.502, "step": 2976 }, { "epoch": 32.0, "grad_norm": 1.4624749422073364, "learning_rate": 3.4000000000000007e-05, "loss": 0.0582, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9015544041450777, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.87, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8783382789317508, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8654970760233918, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.046220194548368454, "eval_overall_accuracy": 0.9849108367626886, "eval_overall_f1": 0.9142156862745099, "eval_overall_precision": 0.8987951807228916, "eval_overall_recall": 0.9301745635910225, "eval_runtime": 0.2405, "eval_samples_per_second": 706.758, "eval_steps_per_second": 12.472, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.7480071783065796, "learning_rate": 3.35e-05, "loss": 0.058, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.8923076923076922, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8529411764705882, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8849557522123894, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8670520231213873, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.046276405453681946, "eval_overall_accuracy": 0.9843621399176955, "eval_overall_f1": 0.9146341463414634, "eval_overall_precision": 0.8949880668257757, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2417, "eval_samples_per_second": 703.371, "eval_steps_per_second": 12.412, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.6384849548339844, "learning_rate": 3.3e-05, "loss": 0.0587, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.9015544041450777, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.87, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.883720930232558, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8539325842696629, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04486535117030144, "eval_overall_accuracy": 0.9843621399176955, "eval_overall_f1": 0.9172749391727494, "eval_overall_precision": 0.8954869358669834, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2415, "eval_samples_per_second": 703.893, "eval_steps_per_second": 12.422, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.5088446736335754, "learning_rate": 3.2500000000000004e-05, "loss": 0.056, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.8969072164948454, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8613861386138614, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8830409356725145, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8579545454545454, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.044350359588861465, "eval_overall_accuracy": 0.9843621399176955, "eval_overall_f1": 0.9159561510353227, "eval_overall_precision": 0.8952380952380953, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2412, "eval_samples_per_second": 704.892, "eval_steps_per_second": 12.439, "step": 3360 }, { "epoch": 36.0, "grad_norm": 1.4163638353347778, "learning_rate": 3.2000000000000005e-05, "loss": 0.0573, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.8969072164948454, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8613861386138614, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8882352941176471, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.867816091954023, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.044320087879896164, "eval_overall_accuracy": 0.984636488340192, "eval_overall_f1": 0.9181929181929182, "eval_overall_precision": 0.8995215311004785, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2413, "eval_samples_per_second": 704.491, "eval_steps_per_second": 12.432, "step": 3456 }, { "epoch": 37.0, "grad_norm": 1.3629268407821655, "learning_rate": 3.15e-05, "loss": 0.0538, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.8958333333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8686868686868687, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.8830409356725145, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8579545454545454, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04334717243909836, "eval_overall_accuracy": 0.9854595336076818, "eval_overall_f1": 0.9157509157509157, "eval_overall_precision": 0.8971291866028708, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2417, "eval_samples_per_second": 703.413, "eval_steps_per_second": 12.413, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.3516825735569, "learning_rate": 3.1e-05, "loss": 0.0533, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.8958333333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8686868686868687, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.8823529411764707, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8620689655172413, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.043126873672008514, "eval_overall_accuracy": 0.9849108367626886, "eval_overall_f1": 0.9155446756425949, "eval_overall_precision": 0.8990384615384616, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.242, "eval_samples_per_second": 702.567, "eval_steps_per_second": 12.398, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.826287567615509, "learning_rate": 3.05e-05, "loss": 0.0513, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.8958333333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8686868686868687, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.8908554572271385, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8728323699421965, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04249382019042969, "eval_overall_accuracy": 0.9860082304526749, "eval_overall_f1": 0.9191176470588236, "eval_overall_precision": 0.9036144578313253, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2413, "eval_samples_per_second": 704.593, "eval_steps_per_second": 12.434, "step": 3744 }, { "epoch": 40.0, "grad_norm": 1.4402703046798706, "learning_rate": 3e-05, "loss": 0.0512, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.8900523560209423, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8673469387755102, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.8790560471976401, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.861271676300578, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.043152716010808945, "eval_overall_accuracy": 0.9849108367626886, "eval_overall_f1": 0.9117647058823528, "eval_overall_precision": 0.8963855421686747, "eval_overall_recall": 0.9276807980049875, "eval_runtime": 0.241, "eval_samples_per_second": 705.333, "eval_steps_per_second": 12.447, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.5191991925239563, "learning_rate": 2.95e-05, "loss": 0.051, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.8900523560209423, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8673469387755102, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.8882352941176471, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.867816091954023, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0419062040746212, "eval_overall_accuracy": 0.9854595336076818, "eval_overall_f1": 0.9166666666666666, "eval_overall_precision": 0.9012048192771084, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2415, "eval_samples_per_second": 704.059, "eval_steps_per_second": 12.425, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.8008220195770264, "learning_rate": 2.9e-05, "loss": 0.0504, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.8900523560209423, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8673469387755102, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.8856304985337244, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8628571428571429, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.041928309947252274, "eval_overall_accuracy": 0.9851851851851852, "eval_overall_f1": 0.9155446756425949, "eval_overall_precision": 0.8990384615384616, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2429, "eval_samples_per_second": 699.741, "eval_steps_per_second": 12.348, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.6116604804992676, "learning_rate": 2.8499999999999998e-05, "loss": 0.0467, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.8900523560209423, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8673469387755102, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.8908554572271385, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8728323699421965, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.042027004063129425, "eval_overall_accuracy": 0.9851851851851852, "eval_overall_f1": 0.9177914110429448, "eval_overall_precision": 0.9033816425120773, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2423, "eval_samples_per_second": 701.558, "eval_steps_per_second": 12.38, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.5427521467208862, "learning_rate": 2.8000000000000003e-05, "loss": 0.0495, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.900523560209424, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8775510204081632, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.8823529411764707, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8620689655172413, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.042836010456085205, "eval_overall_accuracy": 0.9849108367626886, "eval_overall_f1": 0.9155446756425949, "eval_overall_precision": 0.8990384615384616, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2409, "eval_samples_per_second": 705.729, "eval_steps_per_second": 12.454, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.8470682501792908, "learning_rate": 2.7500000000000004e-05, "loss": 0.0481, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.8958333333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8686868686868687, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.8934911242603549, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.877906976744186, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.041298165917396545, "eval_overall_accuracy": 0.9857338820301783, "eval_overall_f1": 0.9202453987730063, "eval_overall_precision": 0.9057971014492754, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2424, "eval_samples_per_second": 701.373, "eval_steps_per_second": 12.377, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.35606712102890015, "learning_rate": 2.7000000000000002e-05, "loss": 0.045, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.90625, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.890207715133531, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8771929824561403, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04163467139005661, "eval_overall_accuracy": 0.9860082304526749, "eval_overall_f1": 0.9202453987730063, "eval_overall_precision": 0.9057971014492754, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2413, "eval_samples_per_second": 704.623, "eval_steps_per_second": 12.435, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.15657056868076324, "learning_rate": 2.6500000000000004e-05, "loss": 0.0459, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.900523560209424, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8775510204081632, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.896755162241888, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8786127167630058, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.041414789855480194, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9226993865030675, "eval_overall_precision": 0.9082125603864735, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2414, "eval_samples_per_second": 704.332, "eval_steps_per_second": 12.429, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.42030104994773865, "learning_rate": 2.6000000000000002e-05, "loss": 0.0452, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9015544041450777, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.87, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8862275449101796, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8809523809523809, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04239359870553017, "eval_overall_accuracy": 0.9857338820301783, "eval_overall_f1": 0.917589175891759, "eval_overall_precision": 0.9053398058252428, "eval_overall_recall": 0.9301745635910225, "eval_runtime": 0.2454, "eval_samples_per_second": 692.721, "eval_steps_per_second": 12.224, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.6118212938308716, "learning_rate": 2.5500000000000003e-05, "loss": 0.0451, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9109947643979057, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8877551020408163, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9053254437869822, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8895348837209303, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04077785089612007, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9287469287469288, "eval_overall_precision": 0.9152542372881356, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2464, "eval_samples_per_second": 689.975, "eval_steps_per_second": 12.176, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.44020506739616394, "learning_rate": 2.5e-05, "loss": 0.0432, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.8958333333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8686868686868687, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.8994082840236686, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8837209302325582, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.041312020272016525, "eval_overall_accuracy": 0.9865569272976681, "eval_overall_f1": 0.9226993865030675, "eval_overall_precision": 0.9082125603864735, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2471, "eval_samples_per_second": 688.097, "eval_steps_per_second": 12.143, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.7177369594573975, "learning_rate": 2.45e-05, "loss": 0.0454, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.9015544041450777, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.87, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8955223880597014, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8875739644970414, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04165731742978096, "eval_overall_accuracy": 0.9865569272976681, "eval_overall_f1": 0.9213759213759214, "eval_overall_precision": 0.9079903147699758, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2465, "eval_samples_per_second": 689.691, "eval_steps_per_second": 12.171, "step": 4896 }, { "epoch": 52.0, "grad_norm": 1.029579758644104, "learning_rate": 2.4e-05, "loss": 0.044, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.894736842105263, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8762886597938144, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.8941176470588235, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8735632183908046, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04128466919064522, "eval_overall_accuracy": 0.9860082304526749, "eval_overall_f1": 0.9202453987730063, "eval_overall_precision": 0.9057971014492754, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2469, "eval_samples_per_second": 688.438, "eval_steps_per_second": 12.149, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.4525042474269867, "learning_rate": 2.35e-05, "loss": 0.0438, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.90625, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8994082840236686, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8837209302325582, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04248586297035217, "eval_overall_accuracy": 0.9865569272976681, "eval_overall_f1": 0.9251533742331289, "eval_overall_precision": 0.9106280193236715, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.246, "eval_samples_per_second": 690.925, "eval_steps_per_second": 12.193, "step": 5088 }, { "epoch": 54.0, "grad_norm": 1.3873966932296753, "learning_rate": 2.3000000000000003e-05, "loss": 0.0426, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.8854166666666665, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8585858585858586, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.8908554572271385, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8728323699421965, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04114874079823494, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9166666666666666, "eval_overall_precision": 0.9012048192771084, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2486, "eval_samples_per_second": 683.895, "eval_steps_per_second": 12.069, "step": 5184 }, { "epoch": 55.0, "grad_norm": 1.1131951808929443, "learning_rate": 2.25e-05, "loss": 0.0423, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.8958333333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8686868686868687, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.8842729970326408, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8713450292397661, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.040763795375823975, "eval_overall_accuracy": 0.9857338820301783, "eval_overall_f1": 0.9153374233128836, "eval_overall_precision": 0.9009661835748792, "eval_overall_recall": 0.9301745635910225, "eval_runtime": 0.2466, "eval_samples_per_second": 689.482, "eval_steps_per_second": 12.167, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.7127313017845154, "learning_rate": 2.2000000000000003e-05, "loss": 0.042, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.900523560209424, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8775510204081632, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.9053254437869822, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8895348837209303, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.040777694433927536, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9262899262899262, "eval_overall_precision": 0.9128329297820823, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2468, "eval_samples_per_second": 688.951, "eval_steps_per_second": 12.158, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.6182370185852051, "learning_rate": 2.15e-05, "loss": 0.0433, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.8969072164948454, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8613861386138614, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8862275449101796, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8809523809523809, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.041250262409448624, "eval_overall_accuracy": 0.9854595336076818, "eval_overall_f1": 0.9164619164619164, "eval_overall_precision": 0.9031476997578692, "eval_overall_recall": 0.9301745635910225, "eval_runtime": 0.2474, "eval_samples_per_second": 687.163, "eval_steps_per_second": 12.126, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.5789722204208374, "learning_rate": 2.1e-05, "loss": 0.0433, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9109947643979057, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8877551020408163, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8869047619047619, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8764705882352941, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04161127284169197, "eval_overall_accuracy": 0.9857338820301783, "eval_overall_f1": 0.9200492004920048, "eval_overall_precision": 0.9077669902912622, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2464, "eval_samples_per_second": 689.936, "eval_steps_per_second": 12.175, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.3667323887348175, "learning_rate": 2.05e-05, "loss": 0.0391, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.8969072164948454, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8613861386138614, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8862275449101796, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8809523809523809, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04131270572543144, "eval_overall_accuracy": 0.9854595336076818, "eval_overall_f1": 0.9164619164619164, "eval_overall_precision": 0.9031476997578692, "eval_overall_recall": 0.9301745635910225, "eval_runtime": 0.2465, "eval_samples_per_second": 689.522, "eval_steps_per_second": 12.168, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.9613834619522095, "learning_rate": 2e-05, "loss": 0.0403, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.9025641025641027, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8627450980392157, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8922155688622753, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8869047619047619, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04109574481844902, "eval_overall_accuracy": 0.9857338820301783, "eval_overall_f1": 0.9202453987730063, "eval_overall_precision": 0.9057971014492754, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.246, "eval_samples_per_second": 690.944, "eval_steps_per_second": 12.193, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.6008909940719604, "learning_rate": 1.9500000000000003e-05, "loss": 0.0397, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.9052631578947367, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8865979381443299, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.9080118694362017, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8947368421052632, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.039610207080841064, "eval_overall_accuracy": 0.9865569272976681, "eval_overall_f1": 0.9285714285714285, "eval_overall_precision": 0.9172749391727494, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2465, "eval_samples_per_second": 689.529, "eval_steps_per_second": 12.168, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.6965606808662415, "learning_rate": 1.9e-05, "loss": 0.0384, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.9015544041450777, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.87, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8988095238095238, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.888235294117647, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.040546514093875885, "eval_overall_accuracy": 0.9860082304526749, "eval_overall_f1": 0.9238329238329237, "eval_overall_precision": 0.910411622276029, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.246, "eval_samples_per_second": 690.94, "eval_steps_per_second": 12.193, "step": 5952 }, { "epoch": 63.0, "grad_norm": 1.4562734365463257, "learning_rate": 1.85e-05, "loss": 0.0388, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.9072164948453608, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8712871287128713, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.8982035928143712, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8928571428571429, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0406072661280632, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9249692496924969, "eval_overall_precision": 0.912621359223301, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2453, "eval_samples_per_second": 693.047, "eval_steps_per_second": 12.23, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.8412647843360901, "learning_rate": 1.8e-05, "loss": 0.0372, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.9109947643979057, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8877551020408163, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9107142857142857, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0397164486348629, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9310344827586207, "eval_overall_precision": 0.9197080291970803, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.248, "eval_samples_per_second": 685.384, "eval_steps_per_second": 12.095, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.5504492521286011, "learning_rate": 1.75e-05, "loss": 0.0368, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.911917098445596, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.88, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9047619047619048, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8941176470588236, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.040799230337142944, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9287469287469288, "eval_overall_precision": 0.9152542372881356, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2461, "eval_samples_per_second": 690.647, "eval_steps_per_second": 12.188, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.5533114075660706, "learning_rate": 1.7000000000000003e-05, "loss": 0.0385, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.90625, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9074626865671641, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8994082840236687, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.039892882108688354, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9285714285714285, "eval_overall_precision": 0.9172749391727494, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2466, "eval_samples_per_second": 689.258, "eval_steps_per_second": 12.163, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.2566806972026825, "learning_rate": 1.65e-05, "loss": 0.0366, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9072164948453608, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8712871287128713, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9041916167664671, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8988095238095238, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.040724191814661026, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9262899262899262, "eval_overall_precision": 0.9128329297820823, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2457, "eval_samples_per_second": 691.771, "eval_steps_per_second": 12.208, "step": 6432 }, { "epoch": 68.0, "grad_norm": 1.0337544679641724, "learning_rate": 1.6000000000000003e-05, "loss": 0.0358, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03972390666604042, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9369592088998765, "eval_overall_precision": 0.928921568627451, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2455, "eval_samples_per_second": 692.515, "eval_steps_per_second": 12.221, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.9264416694641113, "learning_rate": 1.55e-05, "loss": 0.0366, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.9319371727748691, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9096385542168675, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9096385542168675, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03978828340768814, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9344870210135972, "eval_overall_precision": 0.9264705882352942, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2408, "eval_samples_per_second": 706.091, "eval_steps_per_second": 12.46, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.6838388442993164, "learning_rate": 1.5e-05, "loss": 0.0362, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9047619047619048, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03955680504441261, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9321824907521579, "eval_overall_precision": 0.9219512195121952, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2413, "eval_samples_per_second": 704.427, "eval_steps_per_second": 12.431, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.6758135557174683, "learning_rate": 1.45e-05, "loss": 0.0353, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9253731343283581, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9171597633136095, "eval_ORGANIZATION_recall": 0.9337349397590361, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.039997898042201996, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9445129469790383, "eval_overall_precision": 0.9341463414634147, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2465, "eval_samples_per_second": 689.659, "eval_steps_per_second": 12.17, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.5128298997879028, "learning_rate": 1.4000000000000001e-05, "loss": 0.0358, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03966869041323662, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9333333333333333, "eval_overall_precision": 0.9242053789731052, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.246, "eval_samples_per_second": 690.926, "eval_steps_per_second": 12.193, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.7278727293014526, "learning_rate": 1.3500000000000001e-05, "loss": 0.0333, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.039260078221559525, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9358024691358026, "eval_overall_precision": 0.9266503667481663, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2467, "eval_samples_per_second": 689.174, "eval_steps_per_second": 12.162, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.237697571516037, "learning_rate": 1.3000000000000001e-05, "loss": 0.0338, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9096385542168675, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9096385542168675, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03926811367273331, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9320148331273177, "eval_overall_precision": 0.9240196078431373, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2471, "eval_samples_per_second": 688.111, "eval_steps_per_second": 12.143, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.649970531463623, "learning_rate": 1.25e-05, "loss": 0.0368, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.039759960025548935, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9407407407407408, "eval_overall_precision": 0.9315403422982885, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2461, "eval_samples_per_second": 690.875, "eval_steps_per_second": 12.192, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.20745112001895905, "learning_rate": 1.2e-05, "loss": 0.034, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03980717062950134, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9344870210135972, "eval_overall_precision": 0.9264705882352942, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2461, "eval_samples_per_second": 690.818, "eval_steps_per_second": 12.191, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.31113100051879883, "learning_rate": 1.1500000000000002e-05, "loss": 0.0358, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03914351761341095, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9344870210135972, "eval_overall_precision": 0.9264705882352942, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2476, "eval_samples_per_second": 686.641, "eval_steps_per_second": 12.117, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.7353150248527527, "learning_rate": 1.1000000000000001e-05, "loss": 0.0343, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9096385542168675, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9096385542168675, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03905315324664116, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9308641975308642, "eval_overall_precision": 0.921760391198044, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2472, "eval_samples_per_second": 687.639, "eval_steps_per_second": 12.135, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.30361926555633545, "learning_rate": 1.05e-05, "loss": 0.0338, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.906906906906907, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9041916167664671, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.038806237280368805, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9308641975308642, "eval_overall_precision": 0.921760391198044, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2465, "eval_samples_per_second": 689.665, "eval_steps_per_second": 12.171, "step": 7584 }, { "epoch": 80.0, "grad_norm": 1.1131361722946167, "learning_rate": 1e-05, "loss": 0.0343, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9107142857142857, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03866330534219742, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.932349323493235, "eval_overall_precision": 0.9199029126213593, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2457, "eval_samples_per_second": 691.859, "eval_steps_per_second": 12.209, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.6379571557044983, "learning_rate": 9.5e-06, "loss": 0.0345, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.9374999999999999, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03968603536486626, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9395807644882862, "eval_overall_precision": 0.9292682926829269, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2455, "eval_samples_per_second": 692.572, "eval_steps_per_second": 12.222, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.5990679264068604, "learning_rate": 9e-06, "loss": 0.0332, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9374999999999999, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9096385542168675, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9096385542168675, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.040456149727106094, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9358024691358026, "eval_overall_precision": 0.9266503667481663, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.247, "eval_samples_per_second": 688.124, "eval_steps_per_second": 12.143, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.177412748336792, "learning_rate": 8.500000000000002e-06, "loss": 0.0341, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9063444108761329, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9090909090909091, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.040099721401929855, "eval_overall_accuracy": 0.9865569272976681, "eval_overall_f1": 0.9295426452410381, "eval_overall_precision": 0.9215686274509803, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2455, "eval_samples_per_second": 692.361, "eval_steps_per_second": 12.218, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.37546971440315247, "learning_rate": 8.000000000000001e-06, "loss": 0.0323, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.040154702961444855, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9358024691358026, "eval_overall_precision": 0.9266503667481663, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2458, "eval_samples_per_second": 691.483, "eval_steps_per_second": 12.203, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.6602434515953064, "learning_rate": 7.5e-06, "loss": 0.0331, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03959035128355026, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9333333333333333, "eval_overall_precision": 0.9242053789731052, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2456, "eval_samples_per_second": 692.054, "eval_steps_per_second": 12.213, "step": 8160 }, { "epoch": 86.0, "grad_norm": 0.2509349286556244, "learning_rate": 7.000000000000001e-06, "loss": 0.0337, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9063444108761329, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9090909090909091, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03948034718632698, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9320148331273177, "eval_overall_precision": 0.9240196078431373, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2472, "eval_samples_per_second": 687.777, "eval_steps_per_second": 12.137, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.661189615726471, "learning_rate": 6.5000000000000004e-06, "loss": 0.0336, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9096385542168675, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9096385542168675, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03930089995265007, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9308641975308642, "eval_overall_precision": 0.921760391198044, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2459, "eval_samples_per_second": 691.262, "eval_steps_per_second": 12.199, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.48950380086898804, "learning_rate": 6e-06, "loss": 0.0331, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03947994112968445, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9346485819975339, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2401, "eval_samples_per_second": 707.944, "eval_steps_per_second": 12.493, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.7067318558692932, "learning_rate": 5.500000000000001e-06, "loss": 0.0336, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9096385542168675, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9096385542168675, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.039586834609508514, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9308641975308642, "eval_overall_precision": 0.921760391198044, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2459, "eval_samples_per_second": 691.413, "eval_steps_per_second": 12.201, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.44989824295043945, "learning_rate": 5e-06, "loss": 0.033, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.039642609655857086, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9346485819975339, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2453, "eval_samples_per_second": 693.022, "eval_steps_per_second": 12.23, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.375193327665329, "learning_rate": 4.5e-06, "loss": 0.033, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03953820466995239, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9333333333333333, "eval_overall_precision": 0.9242053789731052, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2465, "eval_samples_per_second": 689.739, "eval_steps_per_second": 12.172, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.14784856140613556, "learning_rate": 4.000000000000001e-06, "loss": 0.0317, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9374999999999999, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03980843722820282, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9382716049382714, "eval_overall_precision": 0.9290953545232273, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2461, "eval_samples_per_second": 690.671, "eval_steps_per_second": 12.188, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.5254346132278442, "learning_rate": 3.5000000000000004e-06, "loss": 0.0337, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.039689503610134125, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9371146732429101, "eval_overall_precision": 0.926829268292683, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2461, "eval_samples_per_second": 690.671, "eval_steps_per_second": 12.188, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.2744879126548767, "learning_rate": 3e-06, "loss": 0.0324, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03970589116215706, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9371146732429101, "eval_overall_precision": 0.926829268292683, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2461, "eval_samples_per_second": 690.871, "eval_steps_per_second": 12.192, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.8289042115211487, "learning_rate": 2.5e-06, "loss": 0.033, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03949153423309326, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9346485819975339, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2452, "eval_samples_per_second": 693.3, "eval_steps_per_second": 12.235, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.32764148712158203, "learning_rate": 2.0000000000000003e-06, "loss": 0.0309, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.039386630058288574, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9371146732429101, "eval_overall_precision": 0.926829268292683, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2465, "eval_samples_per_second": 689.731, "eval_steps_per_second": 12.172, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.12100930511951447, "learning_rate": 1.5e-06, "loss": 0.0322, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.039491165429353714, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9346485819975339, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2464, "eval_samples_per_second": 689.957, "eval_steps_per_second": 12.176, "step": 9312 }, { "epoch": 98.0, "grad_norm": 0.8115513324737549, "learning_rate": 1.0000000000000002e-06, "loss": 0.033, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03956222161650658, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9371146732429101, "eval_overall_precision": 0.926829268292683, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2455, "eval_samples_per_second": 692.459, "eval_steps_per_second": 12.22, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.5287472605705261, "learning_rate": 5.000000000000001e-07, "loss": 0.0318, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0395088866353035, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9346485819975339, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2463, "eval_samples_per_second": 690.259, "eval_steps_per_second": 12.181, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.527046263217926, "learning_rate": 0.0, "loss": 0.0327, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.03950924053788185, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9346485819975339, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2454, "eval_samples_per_second": 692.809, "eval_steps_per_second": 12.226, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 3781956116772078.0, "train_loss": 0.08238348955909411, "train_runtime": 545.9152, "train_samples_per_second": 280.446, "train_steps_per_second": 17.585 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 3781956116772078.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }