{ "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": 1.5395334959030151, "learning_rate": 4.9500000000000004e-05, "loss": 0.8685, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.0, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.0, "eval_LOCATION_recall": 0.0, "eval_ORGANIZATION_f1": 0.21481481481481482, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.25892857142857145, "eval_ORGANIZATION_recall": 0.18354430379746836, "eval_PERSON_f1": 0.27802690582959644, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.31313131313131315, "eval_PERSON_recall": 0.25, "eval_loss": 0.40017521381378174, "eval_overall_accuracy": 0.8600054899807851, "eval_overall_f1": 0.19639934533551553, "eval_overall_precision": 0.28169014084507044, "eval_overall_recall": 0.1507537688442211, "eval_runtime": 0.2917, "eval_samples_per_second": 582.696, "eval_steps_per_second": 10.283, "step": 96 }, { "epoch": 2.0, "grad_norm": 0.7787103652954102, "learning_rate": 4.9e-05, "loss": 0.3663, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.3383458646616541, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.3, "eval_LOCATION_recall": 0.3879310344827586, "eval_ORGANIZATION_f1": 0.5501618122977346, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.5629139072847682, "eval_ORGANIZATION_recall": 0.5379746835443038, "eval_PERSON_f1": 0.6965517241379311, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.608433734939759, "eval_PERSON_recall": 0.8145161290322581, "eval_loss": 0.2149091213941574, "eval_overall_accuracy": 0.9286302497941257, "eval_overall_f1": 0.5341040462427745, "eval_overall_precision": 0.49464668094218417, "eval_overall_recall": 0.5804020100502513, "eval_runtime": 0.3, "eval_samples_per_second": 566.7, "eval_steps_per_second": 10.001, "step": 192 }, { "epoch": 3.0, "grad_norm": 0.8174079060554504, "learning_rate": 4.85e-05, "loss": 0.2004, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.7929515418502202, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8108108108108109, "eval_LOCATION_recall": 0.7758620689655172, "eval_ORGANIZATION_f1": 0.7529411764705883, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.7032967032967034, "eval_ORGANIZATION_recall": 0.810126582278481, "eval_PERSON_f1": 0.9384615384615386, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.8970588235294118, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.10894648730754852, "eval_overall_accuracy": 0.9673346143288498, "eval_overall_f1": 0.8222490931076178, "eval_overall_precision": 0.7925407925407926, "eval_overall_recall": 0.8542713567839196, "eval_runtime": 0.2993, "eval_samples_per_second": 567.918, "eval_steps_per_second": 10.022, "step": 288 }, { "epoch": 4.0, "grad_norm": 0.8832339644432068, "learning_rate": 4.8e-05, "loss": 0.1345, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.8392156862745098, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.7697841726618705, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.8297213622291021, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.8121212121212121, "eval_ORGANIZATION_recall": 0.8481012658227848, "eval_PERSON_f1": 0.9685039370078741, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9461538461538461, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08079170435667038, "eval_overall_accuracy": 0.9730990941531704, "eval_overall_f1": 0.8750000000000001, "eval_overall_precision": 0.8387096774193549, "eval_overall_recall": 0.914572864321608, "eval_runtime": 0.2975, "eval_samples_per_second": 571.443, "eval_steps_per_second": 10.084, "step": 384 }, { "epoch": 5.0, "grad_norm": 0.6509011387825012, "learning_rate": 4.75e-05, "loss": 0.1084, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.8820960698689956, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8938053097345132, "eval_LOCATION_recall": 0.8706896551724138, "eval_ORGANIZATION_f1": 0.89937106918239, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.89375, "eval_ORGANIZATION_recall": 0.9050632911392406, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.055279430001974106, "eval_overall_accuracy": 0.9835300576447983, "eval_overall_f1": 0.9197994987468672, "eval_overall_precision": 0.9175, "eval_overall_recall": 0.9221105527638191, "eval_runtime": 0.2983, "eval_samples_per_second": 569.966, "eval_steps_per_second": 10.058, "step": 480 }, { "epoch": 6.0, "grad_norm": 0.7613300681114197, "learning_rate": 4.7e-05, "loss": 0.0959, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.8861788617886178, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8384615384615385, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.897196261682243, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.8834355828220859, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0483161136507988, "eval_overall_accuracy": 0.9854515509195718, "eval_overall_f1": 0.9204406364749081, "eval_overall_precision": 0.8973747016706444, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2997, "eval_samples_per_second": 567.283, "eval_steps_per_second": 10.011, "step": 576 }, { "epoch": 7.0, "grad_norm": 0.9935237169265747, "learning_rate": 4.6500000000000005e-05, "loss": 0.0827, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.920353982300885, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9454545454545454, "eval_LOCATION_recall": 0.896551724137931, "eval_ORGANIZATION_f1": 0.8982035928143712, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.8522727272727273, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04283222183585167, "eval_overall_accuracy": 0.987098545155092, "eval_overall_f1": 0.9306930693069306, "eval_overall_precision": 0.9170731707317074, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.3003, "eval_samples_per_second": 566.104, "eval_steps_per_second": 9.99, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.3799850642681122, "learning_rate": 4.600000000000001e-05, "loss": 0.08, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.9184549356223175, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9145299145299145, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.9249999999999999, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9135802469135802, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03813660517334938, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9401496259351622, "eval_overall_precision": 0.9331683168316832, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.298, "eval_samples_per_second": 570.551, "eval_steps_per_second": 10.069, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.45304006338119507, "learning_rate": 4.55e-05, "loss": 0.0729, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.9045643153526971, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.872, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9283489096573209, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9141104294478528, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04856168106198311, "eval_overall_accuracy": 0.9843535547625584, "eval_overall_f1": 0.9384236453201971, "eval_overall_precision": 0.9202898550724637, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2957, "eval_samples_per_second": 574.919, "eval_steps_per_second": 10.146, "step": 864 }, { "epoch": 10.0, "grad_norm": 0.9706667065620422, "learning_rate": 4.5e-05, "loss": 0.0635, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.8806584362139918, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.84251968503937, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.9249999999999999, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9135802469135802, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.043479882180690765, "eval_overall_accuracy": 0.9857260499588252, "eval_overall_f1": 0.9287469287469288, "eval_overall_precision": 0.9086538461538461, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.3019, "eval_samples_per_second": 563.178, "eval_steps_per_second": 9.938, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.4734399914741516, "learning_rate": 4.4500000000000004e-05, "loss": 0.0614, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.9016393442622951, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.859375, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9155844155844156, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.94, "eval_ORGANIZATION_recall": 0.8924050632911392, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.042100898921489716, "eval_overall_accuracy": 0.9865495470765853, "eval_overall_f1": 0.9326683291770573, "eval_overall_precision": 0.9257425742574258, "eval_overall_recall": 0.9396984924623115, "eval_runtime": 0.2972, "eval_samples_per_second": 572.047, "eval_steps_per_second": 10.095, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.6457267999649048, "learning_rate": 4.4000000000000006e-05, "loss": 0.0561, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.9075630252100839, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8852459016393442, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9294871794871796, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9415584415584416, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03181840106844902, "eval_overall_accuracy": 0.9881965413121054, "eval_overall_f1": 0.94, "eval_overall_precision": 0.9353233830845771, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.3016, "eval_samples_per_second": 563.569, "eval_steps_per_second": 9.945, "step": 1152 }, { "epoch": 13.0, "grad_norm": 0.32204386591911316, "learning_rate": 4.35e-05, "loss": 0.0546, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9400630914826499, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9371069182389937, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.030503783375024796, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9526184538653367, "eval_overall_precision": 0.9455445544554455, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2969, "eval_samples_per_second": 572.544, "eval_steps_per_second": 10.104, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.4022064507007599, "learning_rate": 4.3e-05, "loss": 0.0475, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.927038626609442, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9230769230769231, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.939297124600639, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9483870967741935, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03484940528869629, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9484276729559749, "eval_overall_precision": 0.9496221662468514, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2966, "eval_samples_per_second": 573.184, "eval_steps_per_second": 10.115, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.5112974047660828, "learning_rate": 4.25e-05, "loss": 0.0464, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9363057324840764, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9423076923076923, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03380124270915985, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9525, "eval_overall_precision": 0.9477611940298507, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2996, "eval_samples_per_second": 567.467, "eval_steps_per_second": 10.014, "step": 1440 }, { "epoch": 16.0, "grad_norm": 0.8939520716667175, "learning_rate": 4.2e-05, "loss": 0.0428, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9430379746835443, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9430379746835443, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03173217177391052, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.951310861423221, "eval_overall_precision": 0.9454094292803971, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2961, "eval_samples_per_second": 574.17, "eval_steps_per_second": 10.132, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.28312012553215027, "learning_rate": 4.15e-05, "loss": 0.0417, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9329073482428115, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9419354838709677, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.032240577042102814, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.951188986232791, "eval_overall_precision": 0.9476309226932669, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.295, "eval_samples_per_second": 576.34, "eval_steps_per_second": 10.171, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.5611312389373779, "learning_rate": 4.1e-05, "loss": 0.0388, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9490445859872612, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9551282051282052, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.033160652965307236, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.957286432160804, "eval_overall_precision": 0.957286432160804, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2986, "eval_samples_per_second": 569.287, "eval_steps_per_second": 10.046, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.3558262288570404, "learning_rate": 4.05e-05, "loss": 0.0366, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.9145299145299146, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9067796610169492, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.9308176100628932, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.925, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.031707994639873505, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9426433915211971, "eval_overall_precision": 0.9356435643564357, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2979, "eval_samples_per_second": 570.71, "eval_steps_per_second": 10.071, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.4148123562335968, "learning_rate": 4e-05, "loss": 0.0373, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.9237288135593221, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9083333333333333, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9363057324840764, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9423076923076923, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03521905094385147, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9475, "eval_overall_precision": 0.9427860696517413, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2972, "eval_samples_per_second": 571.97, "eval_steps_per_second": 10.094, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.5516610741615295, "learning_rate": 3.9500000000000005e-05, "loss": 0.0363, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.9372384937238494, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9105691056910569, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03482896089553833, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9550000000000001, "eval_overall_precision": 0.9502487562189055, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2984, "eval_samples_per_second": 569.642, "eval_steps_per_second": 10.053, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.5190480947494507, "learning_rate": 3.9000000000000006e-05, "loss": 0.0333, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9551282051282052, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9675324675324676, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03263789415359497, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9610062893081761, "eval_overall_precision": 0.9622166246851386, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2978, "eval_samples_per_second": 570.893, "eval_steps_per_second": 10.075, "step": 2112 }, { "epoch": 23.0, "grad_norm": 0.2496182769536972, "learning_rate": 3.85e-05, "loss": 0.0334, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9385113268608414, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9602649006622517, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03321027755737305, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.952020202020202, "eval_overall_precision": 0.9568527918781726, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2992, "eval_samples_per_second": 568.175, "eval_steps_per_second": 10.027, "step": 2208 }, { "epoch": 24.0, "grad_norm": 0.2451947033405304, "learning_rate": 3.8e-05, "loss": 0.0311, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9389067524115756, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.954248366013072, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03355176001787186, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9521410579345088, "eval_overall_precision": 0.9545454545454546, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2987, "eval_samples_per_second": 569.101, "eval_steps_per_second": 10.043, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.8237476944923401, "learning_rate": 3.7500000000000003e-05, "loss": 0.0291, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.9142857142857144, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8682170542635659, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04588455706834793, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9467162329615861, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2981, "eval_samples_per_second": 570.205, "eval_steps_per_second": 10.062, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.2971176207065582, "learning_rate": 3.7e-05, "loss": 0.0309, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9487179487179486, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.940677966101695, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9487179487179488, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.961038961038961, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03225880116224289, "eval_overall_accuracy": 0.9923140269009059, "eval_overall_f1": 0.9596977329974811, "eval_overall_precision": 0.9621212121212122, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2966, "eval_samples_per_second": 573.153, "eval_steps_per_second": 10.114, "step": 2496 }, { "epoch": 27.0, "grad_norm": 0.9443634152412415, "learning_rate": 3.65e-05, "loss": 0.03, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9354838709677419, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9539473684210527, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03374403342604637, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9508196721311476, "eval_overall_precision": 0.9544303797468354, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2968, "eval_samples_per_second": 572.729, "eval_steps_per_second": 10.107, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.22665169835090637, "learning_rate": 3.6e-05, "loss": 0.0251, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9237288135593221, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9083333333333333, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9367088607594937, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9367088607594937, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.0359656922519207, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9463171036204745, "eval_overall_precision": 0.9404466501240695, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2982, "eval_samples_per_second": 570.179, "eval_steps_per_second": 10.062, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.47689852118492126, "learning_rate": 3.55e-05, "loss": 0.0228, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9237288135593221, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9083333333333333, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9430379746835443, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9430379746835443, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04356582462787628, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.9501246882793017, "eval_overall_precision": 0.943069306930693, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.3012, "eval_samples_per_second": 564.366, "eval_steps_per_second": 9.959, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.7457932829856873, "learning_rate": 3.5e-05, "loss": 0.0248, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.9137931034482759, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9137931034482759, "eval_LOCATION_recall": 0.9137931034482759, "eval_ORGANIZATION_f1": 0.9371069182389938, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.93125, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04006287083029747, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.9450000000000001, "eval_overall_precision": 0.9402985074626866, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2984, "eval_samples_per_second": 569.654, "eval_steps_per_second": 10.053, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.4651215076446533, "learning_rate": 3.45e-05, "loss": 0.0238, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03806662932038307, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9583858764186632, "eval_overall_precision": 0.9620253164556962, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2956, "eval_samples_per_second": 575.154, "eval_steps_per_second": 10.15, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.5793745517730713, "learning_rate": 3.4000000000000007e-05, "loss": 0.0213, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9329073482428115, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9419354838709677, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04070785641670227, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9473684210526316, "eval_overall_precision": 0.945, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2979, "eval_samples_per_second": 570.624, "eval_steps_per_second": 10.07, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.6270995736122131, "learning_rate": 3.35e-05, "loss": 0.0222, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9354838709677419, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9539473684210527, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0342167504131794, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.953341740226986, "eval_overall_precision": 0.9569620253164557, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2985, "eval_samples_per_second": 569.574, "eval_steps_per_second": 10.051, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.5759699940681458, "learning_rate": 3.3e-05, "loss": 0.022, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.9237288135593221, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9083333333333333, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9430379746835443, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9430379746835443, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.041709043085575104, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9488139825218478, "eval_overall_precision": 0.9429280397022333, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2973, "eval_samples_per_second": 571.888, "eval_steps_per_second": 10.092, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.6274858117103577, "learning_rate": 3.2500000000000004e-05, "loss": 0.0217, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9152542372881356, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.946031746031746, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9490445859872612, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.040370818227529526, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9486858573216519, "eval_overall_precision": 0.9451371571072319, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2948, "eval_samples_per_second": 576.631, "eval_steps_per_second": 10.176, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.53178870677948, "learning_rate": 3.2000000000000005e-05, "loss": 0.0188, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.9391304347826087, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9473684210526315, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9333333333333332, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9363057324840764, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.039047881960868835, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9485570890840652, "eval_overall_precision": 0.9473684210526315, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.3024, "eval_samples_per_second": 562.228, "eval_steps_per_second": 9.922, "step": 3456 }, { "epoch": 37.0, "grad_norm": 0.7722581028938293, "learning_rate": 3.15e-05, "loss": 0.0212, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.9473684210526316, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9642857142857143, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9526813880126183, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.949685534591195, "eval_ORGANIZATION_recall": 0.9556962025316456, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03522836044430733, "eval_overall_accuracy": 0.9923140269009059, "eval_overall_f1": 0.9596977329974811, "eval_overall_precision": 0.9621212121212122, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2981, "eval_samples_per_second": 570.347, "eval_steps_per_second": 10.065, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.034928902983665466, "learning_rate": 3.1e-05, "loss": 0.0183, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9437229437229439, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9478260869565217, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9463722397476341, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9433962264150944, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.038987092673778534, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9560853199498118, "eval_overall_precision": 0.9548872180451128, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2965, "eval_samples_per_second": 573.299, "eval_steps_per_second": 10.117, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.5672887563705444, "learning_rate": 3.05e-05, "loss": 0.0184, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.9482758620689655, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9482758620689655, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03323935344815254, "eval_overall_accuracy": 0.9925885259401592, "eval_overall_f1": 0.9596977329974811, "eval_overall_precision": 0.9621212121212122, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2983, "eval_samples_per_second": 569.899, "eval_steps_per_second": 10.057, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.5080427527427673, "learning_rate": 3e-05, "loss": 0.0176, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9721115537848605, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9606299212598425, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04094182699918747, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9498746867167919, "eval_overall_precision": 0.9475, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2956, "eval_samples_per_second": 575.099, "eval_steps_per_second": 10.149, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.07212147861719131, "learning_rate": 2.95e-05, "loss": 0.0182, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9523809523809524, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9554140127388535, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03880760073661804, "eval_overall_accuracy": 0.9923140269009059, "eval_overall_f1": 0.9575, "eval_overall_precision": 0.9527363184079602, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2999, "eval_samples_per_second": 566.947, "eval_steps_per_second": 10.005, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.6076759696006775, "learning_rate": 2.9e-05, "loss": 0.0175, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.9327731092436975, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9098360655737705, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03986818715929985, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9551122194513715, "eval_overall_precision": 0.948019801980198, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.298, "eval_samples_per_second": 570.471, "eval_steps_per_second": 10.067, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.21522241830825806, "learning_rate": 2.8499999999999998e-05, "loss": 0.0169, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9437229437229439, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9478260869565217, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9274447949526813, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9245283018867925, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03812630474567413, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.949748743718593, "eval_overall_precision": 0.949748743718593, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2951, "eval_samples_per_second": 576.049, "eval_steps_per_second": 10.166, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.5210365653038025, "learning_rate": 2.8000000000000003e-05, "loss": 0.0166, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9367088607594937, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9173553719008265, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9682539682539683, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.953125, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.039871733635663986, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9526184538653367, "eval_overall_precision": 0.9455445544554455, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.3007, "eval_samples_per_second": 565.291, "eval_steps_per_second": 9.976, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.5988039970397949, "learning_rate": 2.7500000000000004e-05, "loss": 0.015, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.9491525423728815, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9333333333333333, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03624749928712845, "eval_overall_accuracy": 0.9923140269009059, "eval_overall_f1": 0.957286432160804, "eval_overall_precision": 0.957286432160804, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2962, "eval_samples_per_second": 573.84, "eval_steps_per_second": 10.127, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.12654761970043182, "learning_rate": 2.7000000000000002e-05, "loss": 0.0163, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.9198312236286921, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9008264462809917, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04110268875956535, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9488139825218478, "eval_overall_precision": 0.9429280397022333, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2978, "eval_samples_per_second": 570.834, "eval_steps_per_second": 10.074, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.8154733777046204, "learning_rate": 2.6500000000000004e-05, "loss": 0.0163, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.039265744388103485, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.296, "eval_samples_per_second": 574.362, "eval_steps_per_second": 10.136, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.5110801458358765, "learning_rate": 2.6000000000000002e-05, "loss": 0.0136, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04232502728700638, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9547738693467337, "eval_overall_precision": 0.9547738693467337, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2956, "eval_samples_per_second": 575.116, "eval_steps_per_second": 10.149, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.1729801595211029, "learning_rate": 2.5500000000000003e-05, "loss": 0.0147, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9396551724137931, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9396551724137931, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9490445859872612, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9551282051282052, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9838709677419355, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9838709677419355, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.03779783844947815, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9571788413098236, "eval_overall_precision": 0.9595959595959596, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2969, "eval_samples_per_second": 572.49, "eval_steps_per_second": 10.103, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.9957223534584045, "learning_rate": 2.5e-05, "loss": 0.0126, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9490445859872612, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9551282051282052, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.043282754719257355, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2976, "eval_samples_per_second": 571.251, "eval_steps_per_second": 10.081, "step": 4800 }, { "epoch": 51.0, "grad_norm": 1.63614821434021, "learning_rate": 2.45e-05, "loss": 0.0146, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04477479308843613, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2989, "eval_samples_per_second": 568.75, "eval_steps_per_second": 10.037, "step": 4896 }, { "epoch": 52.0, "grad_norm": 0.36906835436820984, "learning_rate": 2.4e-05, "loss": 0.0146, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9367088607594937, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9173553719008265, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.046886004507541656, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9523809523809523, "eval_overall_precision": 0.95, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2983, "eval_samples_per_second": 569.885, "eval_steps_per_second": 10.057, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.3496086001396179, "learning_rate": 2.35e-05, "loss": 0.0119, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9276595744680851, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9159663865546218, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9426751592356688, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9487179487179487, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04656808078289032, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9488139825218478, "eval_overall_precision": 0.9429280397022333, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2973, "eval_samples_per_second": 571.731, "eval_steps_per_second": 10.089, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.98432457447052, "learning_rate": 2.3000000000000003e-05, "loss": 0.0128, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.053139399737119675, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.951310861423221, "eval_overall_precision": 0.9454094292803971, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2968, "eval_samples_per_second": 572.682, "eval_steps_per_second": 10.106, "step": 5184 }, { "epoch": 55.0, "grad_norm": 1.06143057346344, "learning_rate": 2.25e-05, "loss": 0.013, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9358974358974359, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.948051948051948, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04369078949093819, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9510664993726474, "eval_overall_precision": 0.949874686716792, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2975, "eval_samples_per_second": 571.398, "eval_steps_per_second": 10.083, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.19326968491077423, "learning_rate": 2.2000000000000003e-05, "loss": 0.0125, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9389067524115756, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.954248366013072, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04643920063972473, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9485570890840652, "eval_overall_precision": 0.9473684210526315, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2965, "eval_samples_per_second": 573.443, "eval_steps_per_second": 10.12, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.2908190190792084, "learning_rate": 2.15e-05, "loss": 0.012, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.9396551724137931, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9396551724137931, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9333333333333332, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9363057324840764, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04293465241789818, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9498746867167919, "eval_overall_precision": 0.9475, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2998, "eval_samples_per_second": 567.113, "eval_steps_per_second": 10.008, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.03710107132792473, "learning_rate": 2.1e-05, "loss": 0.012, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.939297124600639, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9483870967741935, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04816155135631561, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9498746867167919, "eval_overall_precision": 0.9475, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2969, "eval_samples_per_second": 572.618, "eval_steps_per_second": 10.105, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.24933892488479614, "learning_rate": 2.05e-05, "loss": 0.0123, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9294871794871796, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9415584415584416, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05612450838088989, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9450000000000001, "eval_overall_precision": 0.9402985074626866, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2983, "eval_samples_per_second": 569.802, "eval_steps_per_second": 10.055, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.1557096242904663, "learning_rate": 2e-05, "loss": 0.0117, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05061795935034752, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9547738693467337, "eval_overall_precision": 0.9547738693467337, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2971, "eval_samples_per_second": 572.123, "eval_steps_per_second": 10.096, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.2682698369026184, "learning_rate": 1.9500000000000003e-05, "loss": 0.0115, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.048160944133996964, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9547738693467337, "eval_overall_precision": 0.9547738693467337, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2984, "eval_samples_per_second": 569.698, "eval_steps_per_second": 10.054, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.40208688378334045, "learning_rate": 1.9e-05, "loss": 0.0094, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.939297124600639, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9483870967741935, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05038047209382057, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9547738693467337, "eval_overall_precision": 0.9547738693467337, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2976, "eval_samples_per_second": 571.281, "eval_steps_per_second": 10.081, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.02688334509730339, "learning_rate": 1.85e-05, "loss": 0.0101, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.939297124600639, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9483870967741935, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04784246161580086, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9547738693467337, "eval_overall_precision": 0.9547738693467337, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2979, "eval_samples_per_second": 570.734, "eval_steps_per_second": 10.072, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.07943878322839737, "learning_rate": 1.8e-05, "loss": 0.0096, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04850446432828903, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9571788413098236, "eval_overall_precision": 0.9595959595959596, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2969, "eval_samples_per_second": 572.602, "eval_steps_per_second": 10.105, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.4087267518043518, "learning_rate": 1.75e-05, "loss": 0.0095, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05776742100715637, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.951188986232791, "eval_overall_precision": 0.9476309226932669, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2971, "eval_samples_per_second": 572.201, "eval_steps_per_second": 10.098, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.1425824612379074, "learning_rate": 1.7000000000000003e-05, "loss": 0.0113, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9419354838709677, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9605263157894737, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05492893233895302, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.3009, "eval_samples_per_second": 564.889, "eval_steps_per_second": 9.969, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.02967935986816883, "learning_rate": 1.65e-05, "loss": 0.0088, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.939297124600639, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9483870967741935, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.049883853644132614, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.951188986232791, "eval_overall_precision": 0.9476309226932669, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2989, "eval_samples_per_second": 568.81, "eval_steps_per_second": 10.038, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.5351668000221252, "learning_rate": 1.6000000000000003e-05, "loss": 0.0094, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0438864603638649, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.957286432160804, "eval_overall_precision": 0.957286432160804, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2974, "eval_samples_per_second": 571.629, "eval_steps_per_second": 10.088, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.21662431955337524, "learning_rate": 1.55e-05, "loss": 0.008, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04838420823216438, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9597989949748744, "eval_overall_precision": 0.9597989949748744, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2997, "eval_samples_per_second": 567.269, "eval_steps_per_second": 10.011, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.746005117893219, "learning_rate": 1.5e-05, "loss": 0.0092, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.939297124600639, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9483870967741935, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04468633607029915, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9523809523809523, "eval_overall_precision": 0.95, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2983, "eval_samples_per_second": 569.834, "eval_steps_per_second": 10.056, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.053684864193201065, "learning_rate": 1.45e-05, "loss": 0.008, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9358974358974359, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.948051948051948, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05243850126862526, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9510664993726474, "eval_overall_precision": 0.949874686716792, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2956, "eval_samples_per_second": 575.185, "eval_steps_per_second": 10.15, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.7887480854988098, "learning_rate": 1.4000000000000001e-05, "loss": 0.0084, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04431832954287529, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9560853199498118, "eval_overall_precision": 0.9548872180451128, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2956, "eval_samples_per_second": 575.026, "eval_steps_per_second": 10.148, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.6922429800033569, "learning_rate": 1.3500000000000001e-05, "loss": 0.0085, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9490445859872612, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9551282051282052, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04679226130247116, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9573934837092732, "eval_overall_precision": 0.955, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2979, "eval_samples_per_second": 570.644, "eval_steps_per_second": 10.07, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.31839609146118164, "learning_rate": 1.3000000000000001e-05, "loss": 0.0081, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04726555943489075, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9547738693467337, "eval_overall_precision": 0.9547738693467337, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.297, "eval_samples_per_second": 572.376, "eval_steps_per_second": 10.101, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.17296186089515686, "learning_rate": 1.25e-05, "loss": 0.0087, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.9446808510638298, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9327731092436975, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9389067524115756, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.954248366013072, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04478701949119568, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.3081, "eval_samples_per_second": 551.748, "eval_steps_per_second": 9.737, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.019858483225107193, "learning_rate": 1.2e-05, "loss": 0.0076, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9426751592356688, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9487179487179487, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0498213991522789, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9525, "eval_overall_precision": 0.9477611940298507, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2955, "eval_samples_per_second": 575.326, "eval_steps_per_second": 10.153, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.46187853813171387, "learning_rate": 1.1500000000000002e-05, "loss": 0.0074, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.939297124600639, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9483870967741935, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.047739606350660324, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9523809523809523, "eval_overall_precision": 0.95, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2989, "eval_samples_per_second": 568.755, "eval_steps_per_second": 10.037, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.1097620278596878, "learning_rate": 1.1000000000000001e-05, "loss": 0.008, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9487179487179486, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.940677966101695, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.042885202914476395, "eval_overall_accuracy": 0.9923140269009059, "eval_overall_f1": 0.9623115577889447, "eval_overall_precision": 0.9623115577889447, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2973, "eval_samples_per_second": 571.834, "eval_steps_per_second": 10.091, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.013154038228094578, "learning_rate": 1.05e-05, "loss": 0.0072, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04914812371134758, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.951188986232791, "eval_overall_precision": 0.9476309226932669, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2959, "eval_samples_per_second": 574.45, "eval_steps_per_second": 10.137, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.04681688919663429, "learning_rate": 1e-05, "loss": 0.0066, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9517684887459807, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9673202614379085, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0501355342566967, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9573934837092732, "eval_overall_precision": 0.955, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2946, "eval_samples_per_second": 576.977, "eval_steps_per_second": 10.182, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.43833377957344055, "learning_rate": 9.5e-06, "loss": 0.0068, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04716813564300537, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9548872180451128, "eval_overall_precision": 0.9525, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2995, "eval_samples_per_second": 567.635, "eval_steps_per_second": 10.017, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.1791413277387619, "learning_rate": 9e-06, "loss": 0.0066, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05336508899927139, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9561952440550688, "eval_overall_precision": 0.9526184538653366, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2969, "eval_samples_per_second": 572.633, "eval_steps_per_second": 10.105, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.27082154154777527, "learning_rate": 8.500000000000002e-06, "loss": 0.0069, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04806387424468994, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9560853199498118, "eval_overall_precision": 0.9548872180451128, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2951, "eval_samples_per_second": 576.151, "eval_steps_per_second": 10.167, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.07476259022951126, "learning_rate": 8.000000000000001e-06, "loss": 0.0065, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05209340527653694, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9573934837092732, "eval_overall_precision": 0.955, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2941, "eval_samples_per_second": 577.955, "eval_steps_per_second": 10.199, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.06059735640883446, "learning_rate": 7.5e-06, "loss": 0.0063, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05482008308172226, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9573934837092732, "eval_overall_precision": 0.955, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2961, "eval_samples_per_second": 574.076, "eval_steps_per_second": 10.131, "step": 8160 }, { "epoch": 86.0, "grad_norm": 0.05009327828884125, "learning_rate": 7.000000000000001e-06, "loss": 0.0064, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05267080292105675, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9573934837092732, "eval_overall_precision": 0.955, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2963, "eval_samples_per_second": 573.817, "eval_steps_per_second": 10.126, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.21512138843536377, "learning_rate": 6.5000000000000004e-06, "loss": 0.007, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9490445859872612, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9551282051282052, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0479336678981781, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9573934837092732, "eval_overall_precision": 0.955, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2953, "eval_samples_per_second": 575.737, "eval_steps_per_second": 10.16, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.2677503526210785, "learning_rate": 6e-06, "loss": 0.0062, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9551282051282052, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9675324675324676, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04780548810958862, "eval_overall_accuracy": 0.9917650288223991, "eval_overall_f1": 0.9610062893081761, "eval_overall_precision": 0.9622166246851386, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2974, "eval_samples_per_second": 571.672, "eval_steps_per_second": 10.088, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.7986235022544861, "learning_rate": 5.500000000000001e-06, "loss": 0.0057, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.047579675912857056, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9597989949748744, "eval_overall_precision": 0.9597989949748744, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2943, "eval_samples_per_second": 577.632, "eval_steps_per_second": 10.194, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.5620450973510742, "learning_rate": 5e-06, "loss": 0.0058, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9487179487179488, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.961038961038961, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04754525423049927, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9584905660377357, "eval_overall_precision": 0.9596977329974811, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2939, "eval_samples_per_second": 578.361, "eval_steps_per_second": 10.206, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.1334208995103836, "learning_rate": 4.5e-06, "loss": 0.0066, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05078602954745293, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.958594730238394, "eval_overall_precision": 0.9573934837092731, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2951, "eval_samples_per_second": 576.156, "eval_steps_per_second": 10.167, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.1834694892168045, "learning_rate": 4.000000000000001e-06, "loss": 0.0061, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9446808510638298, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9327731092436975, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9551282051282052, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9675324675324676, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05072775483131409, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9598997493734336, "eval_overall_precision": 0.9575, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2962, "eval_samples_per_second": 573.953, "eval_steps_per_second": 10.129, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.5359702110290527, "learning_rate": 3.5000000000000004e-06, "loss": 0.0065, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9487179487179486, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.940677966101695, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9581993569131834, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9738562091503268, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.047589123249053955, "eval_overall_accuracy": 0.9920395278616525, "eval_overall_f1": 0.9623115577889447, "eval_overall_precision": 0.9623115577889447, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2972, "eval_samples_per_second": 572.082, "eval_steps_per_second": 10.096, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.38933810591697693, "learning_rate": 3e-06, "loss": 0.0065, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04974960908293724, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.958594730238394, "eval_overall_precision": 0.9573934837092731, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2966, "eval_samples_per_second": 573.091, "eval_steps_per_second": 10.113, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.5164521932601929, "learning_rate": 2.5e-06, "loss": 0.007, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04715043306350708, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.958594730238394, "eval_overall_precision": 0.9573934837092731, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.3002, "eval_samples_per_second": 566.322, "eval_steps_per_second": 9.994, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.6935141086578369, "learning_rate": 2.0000000000000003e-06, "loss": 0.0076, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.9487179487179486, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.940677966101695, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9551282051282052, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9675324675324676, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0494760125875473, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9611041405269761, "eval_overall_precision": 0.9598997493734336, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2973, "eval_samples_per_second": 571.835, "eval_steps_per_second": 10.091, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.23275186121463776, "learning_rate": 1.5e-06, "loss": 0.0057, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9487179487179486, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.940677966101695, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9551282051282052, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9675324675324676, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04830334708094597, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9611041405269761, "eval_overall_precision": 0.9598997493734336, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2959, "eval_samples_per_second": 574.573, "eval_steps_per_second": 10.14, "step": 9312 }, { "epoch": 98.0, "grad_norm": 1.0000766515731812, "learning_rate": 1.0000000000000002e-06, "loss": 0.0058, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.952076677316294, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9612903225806452, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.047053366899490356, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.958594730238394, "eval_overall_precision": 0.9573934837092731, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.2956, "eval_samples_per_second": 575.069, "eval_steps_per_second": 10.148, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.2683166563510895, "learning_rate": 5.000000000000001e-07, "loss": 0.0063, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9487179487179486, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.940677966101695, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9551282051282052, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9675324675324676, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.047673843801021576, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9611041405269761, "eval_overall_precision": 0.9598997493734336, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2938, "eval_samples_per_second": 578.7, "eval_steps_per_second": 10.212, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.3831656873226166, "learning_rate": 0.0, "loss": 0.0055, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9487179487179486, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.940677966101695, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9551282051282052, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9675324675324676, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04736623167991638, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9611041405269761, "eval_overall_precision": 0.9598997493734336, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2957, "eval_samples_per_second": 574.889, "eval_steps_per_second": 10.145, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 4293597283743744.0, "train_loss": 0.036266621196021635, "train_runtime": 548.5671, "train_samples_per_second": 279.091, "train_steps_per_second": 17.5 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 4293597283743744.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }