{ "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": 1.0955090522766113, "learning_rate": 4.9500000000000004e-05, "loss": 0.8608, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.03960396039603961, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.25, "eval_LOCATION_recall": 0.021505376344086023, "eval_ORGANIZATION_f1": 0.22727272727272724, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.21505376344086022, "eval_ORGANIZATION_recall": 0.24096385542168675, "eval_PERSON_f1": 0.2815884476534296, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.28888888888888886, "eval_PERSON_recall": 0.2746478873239437, "eval_loss": 0.3943306505680084, "eval_overall_accuracy": 0.8650205761316873, "eval_overall_f1": 0.2219178082191781, "eval_overall_precision": 0.24620060790273557, "eval_overall_recall": 0.20199501246882792, "eval_runtime": 0.2929, "eval_samples_per_second": 580.419, "eval_steps_per_second": 10.243, "step": 96 }, { "epoch": 2.0, "grad_norm": 0.8960049152374268, "learning_rate": 4.9e-05, "loss": 0.3625, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.4050632911392405, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.3333333333333333, "eval_LOCATION_recall": 0.5161290322580645, "eval_ORGANIZATION_f1": 0.6056782334384858, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.6357615894039735, "eval_ORGANIZATION_recall": 0.5783132530120482, "eval_PERSON_f1": 0.6726726726726726, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.5863874345549738, "eval_PERSON_recall": 0.7887323943661971, "eval_loss": 0.21519558131694794, "eval_overall_accuracy": 0.9404663923182441, "eval_overall_f1": 0.5772266065388952, "eval_overall_precision": 0.5267489711934157, "eval_overall_recall": 0.6384039900249376, "eval_runtime": 0.2929, "eval_samples_per_second": 580.324, "eval_steps_per_second": 10.241, "step": 192 }, { "epoch": 3.0, "grad_norm": 0.8648101687431335, "learning_rate": 4.85e-05, "loss": 0.1958, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.7909604519774012, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8333333333333334, "eval_LOCATION_recall": 0.7526881720430108, "eval_ORGANIZATION_f1": 0.7342465753424658, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.6733668341708543, "eval_ORGANIZATION_recall": 0.8072289156626506, "eval_PERSON_f1": 0.9655172413793103, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9459459459459459, "eval_PERSON_recall": 0.9859154929577465, "eval_loss": 0.09694702178239822, "eval_overall_accuracy": 0.9676268861454047, "eval_overall_f1": 0.826923076923077, "eval_overall_precision": 0.7981438515081206, "eval_overall_recall": 0.85785536159601, "eval_runtime": 0.2911, "eval_samples_per_second": 583.932, "eval_steps_per_second": 10.305, "step": 288 }, { "epoch": 4.0, "grad_norm": 1.2493984699249268, "learning_rate": 4.8e-05, "loss": 0.1348, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.7536231884057972, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.6842105263157895, "eval_LOCATION_recall": 0.8387096774193549, "eval_ORGANIZATION_f1": 0.8033240997229917, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.7435897435897436, "eval_ORGANIZATION_recall": 0.8734939759036144, "eval_PERSON_f1": 0.979020979020979, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9722222222222222, "eval_PERSON_recall": 0.9859154929577465, "eval_loss": 0.08365878462791443, "eval_overall_accuracy": 0.9711934156378601, "eval_overall_f1": 0.8501170960187353, "eval_overall_precision": 0.8013245033112583, "eval_overall_recall": 0.9052369077306733, "eval_runtime": 0.2952, "eval_samples_per_second": 575.8, "eval_steps_per_second": 10.161, "step": 384 }, { "epoch": 5.0, "grad_norm": 0.48399850726127625, "learning_rate": 4.75e-05, "loss": 0.1081, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.8633879781420764, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8777777777777778, "eval_LOCATION_recall": 0.8494623655913979, "eval_ORGANIZATION_f1": 0.8563049853372433, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8342857142857143, "eval_ORGANIZATION_recall": 0.8795180722891566, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06440697610378265, "eval_overall_accuracy": 0.9807956104252401, "eval_overall_f1": 0.8985148514851485, "eval_overall_precision": 0.8918918918918919, "eval_overall_recall": 0.9052369077306733, "eval_runtime": 0.2894, "eval_samples_per_second": 587.518, "eval_steps_per_second": 10.368, "step": 480 }, { "epoch": 6.0, "grad_norm": 0.6719767451286316, "learning_rate": 4.7e-05, "loss": 0.0952, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.8195121951219512, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.75, "eval_LOCATION_recall": 0.9032258064516129, "eval_ORGANIZATION_f1": 0.8580060422960725, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8606060606060606, "eval_ORGANIZATION_recall": 0.8554216867469879, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05366777628660202, "eval_overall_accuracy": 0.9824417009602194, "eval_overall_f1": 0.8913308913308914, "eval_overall_precision": 0.8732057416267942, "eval_overall_recall": 0.9102244389027432, "eval_runtime": 0.2926, "eval_samples_per_second": 580.972, "eval_steps_per_second": 10.252, "step": 576 }, { "epoch": 7.0, "grad_norm": 1.178981900215149, "learning_rate": 4.6500000000000005e-05, "loss": 0.0875, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.849740932642487, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.82, "eval_LOCATION_recall": 0.8817204301075269, "eval_ORGANIZATION_f1": 0.888888888888889, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8432432432432433, "eval_ORGANIZATION_recall": 0.9397590361445783, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05959905683994293, "eval_overall_accuracy": 0.9816186556927298, "eval_overall_f1": 0.9106280193236714, "eval_overall_precision": 0.882903981264637, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2883, "eval_samples_per_second": 589.728, "eval_steps_per_second": 10.407, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.6990514993667603, "learning_rate": 4.600000000000001e-05, "loss": 0.073, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.883248730964467, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8365384615384616, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9032258064516129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.88, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05377694219350815, "eval_overall_accuracy": 0.9835390946502057, "eval_overall_f1": 0.9245742092457422, "eval_overall_precision": 0.9026128266033254, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.292, "eval_samples_per_second": 582.272, "eval_steps_per_second": 10.275, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.3934107720851898, "learning_rate": 4.55e-05, "loss": 0.0696, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.8900523560209423, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8673469387755102, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.88955223880597, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8816568047337278, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04754691570997238, "eval_overall_accuracy": 0.9857338820301783, "eval_overall_f1": 0.9185185185185184, "eval_overall_precision": 0.9095354523227384, "eval_overall_recall": 0.9276807980049875, "eval_runtime": 0.2917, "eval_samples_per_second": 582.8, "eval_steps_per_second": 10.285, "step": 864 }, { "epoch": 10.0, "grad_norm": 0.8873964548110962, "learning_rate": 4.5e-05, "loss": 0.064, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9179331306990882, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.052041731774806976, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2881, "eval_samples_per_second": 590.045, "eval_steps_per_second": 10.413, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.5631869435310364, "learning_rate": 4.4500000000000004e-05, "loss": 0.0626, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9169230769230768, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9371069182389937, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.043987125158309937, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9353233830845771, "eval_overall_precision": 0.9330024813895782, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2902, "eval_samples_per_second": 585.729, "eval_steps_per_second": 10.336, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.559025228023529, "learning_rate": 4.4000000000000006e-05, "loss": 0.0554, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.9139784946236559, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9139784946236559, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.9058823529411766, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8850574712643678, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04547671601176262, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9321824907521579, "eval_overall_precision": 0.9219512195121952, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2883, "eval_samples_per_second": 589.693, "eval_steps_per_second": 10.406, "step": 1152 }, { "epoch": 13.0, "grad_norm": 1.0373061895370483, "learning_rate": 4.35e-05, "loss": 0.0548, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.9197860962566845, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9148936170212766, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.9317507418397626, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9181286549707602, "eval_ORGANIZATION_recall": 0.9457831325301205, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04833203926682472, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9443757725587144, "eval_overall_precision": 0.9362745098039216, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2908, "eval_samples_per_second": 584.59, "eval_steps_per_second": 10.316, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.2547779381275177, "learning_rate": 4.3e-05, "loss": 0.0504, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.9247311827956989, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9247311827956989, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.9461077844311376, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9404761904761905, "eval_ORGANIZATION_recall": 0.9518072289156626, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04443012550473213, "eval_overall_accuracy": 0.9895747599451303, "eval_overall_f1": 0.9490683229813663, "eval_overall_precision": 0.9455445544554455, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2897, "eval_samples_per_second": 586.886, "eval_steps_per_second": 10.357, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.5698651075363159, "learning_rate": 4.25e-05, "loss": 0.0494, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.9354838709677419, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9354838709677419, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9317507418397626, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9181286549707602, "eval_ORGANIZATION_recall": 0.9457831325301205, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.042747754603624344, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9480198019801981, "eval_overall_precision": 0.941031941031941, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2871, "eval_samples_per_second": 592.23, "eval_steps_per_second": 10.451, "step": 1440 }, { "epoch": 16.0, "grad_norm": 0.5443034172058105, "learning_rate": 4.2e-05, "loss": 0.0445, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.8989898989898989, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8476190476190476, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.049201611429452896, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9287469287469288, "eval_overall_precision": 0.9152542372881356, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.286, "eval_samples_per_second": 594.461, "eval_steps_per_second": 10.49, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.4349294602870941, "learning_rate": 4.15e-05, "loss": 0.0457, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9397590361445783, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9397590361445783, "eval_ORGANIZATION_recall": 0.9397590361445783, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04454689472913742, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9504950495049505, "eval_overall_precision": 0.9434889434889435, "eval_overall_recall": 0.9576059850374065, "eval_runtime": 0.2825, "eval_samples_per_second": 601.765, "eval_steps_per_second": 10.619, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.2558753192424774, "learning_rate": 4.1e-05, "loss": 0.0433, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9305135951661632, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9333333333333333, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05001579597592354, "eval_overall_accuracy": 0.9895747599451303, "eval_overall_f1": 0.946716232961586, "eval_overall_precision": 0.9408866995073891, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2887, "eval_samples_per_second": 588.934, "eval_steps_per_second": 10.393, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.414883553981781, "learning_rate": 4.05e-05, "loss": 0.0395, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.9304812834224598, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.925531914893617, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9285714285714286, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9176470588235294, "eval_ORGANIZATION_recall": 0.9397590361445783, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04901066794991493, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.946716232961586, "eval_overall_precision": 0.9408866995073891, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2826, "eval_samples_per_second": 601.556, "eval_steps_per_second": 10.616, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.6924071311950684, "learning_rate": 4e-05, "loss": 0.0375, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.9263157894736843, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9072164948453608, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9289940828402367, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9127906976744186, "eval_ORGANIZATION_recall": 0.9457831325301205, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04936488717794418, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9458128078817734, "eval_overall_precision": 0.9343065693430657, "eval_overall_recall": 0.9576059850374065, "eval_runtime": 0.2851, "eval_samples_per_second": 596.276, "eval_steps_per_second": 10.523, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.32161277532577515, "learning_rate": 3.9500000000000005e-05, "loss": 0.0394, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.9319371727748691, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9281437125748503, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9226190476190477, "eval_ORGANIZATION_recall": 0.9337349397590361, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05034565553069115, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9443757725587144, "eval_overall_precision": 0.9362745098039216, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2873, "eval_samples_per_second": 591.81, "eval_steps_per_second": 10.444, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.15593692660331726, "learning_rate": 3.9000000000000006e-05, "loss": 0.0364, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.920353982300885, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9017341040462428, "eval_ORGANIZATION_recall": 0.9397590361445783, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04979310929775238, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.941031941031941, "eval_overall_precision": 0.927360774818402, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2865, "eval_samples_per_second": 593.269, "eval_steps_per_second": 10.469, "step": 2112 }, { "epoch": 23.0, "grad_norm": 1.1374483108520508, "learning_rate": 3.85e-05, "loss": 0.035, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.9109947643979057, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8877551020408163, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.924924924924925, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9221556886227545, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04968388378620148, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9381188118811881, "eval_overall_precision": 0.9312039312039312, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2891, "eval_samples_per_second": 588.072, "eval_steps_per_second": 10.378, "step": 2208 }, { "epoch": 24.0, "grad_norm": 0.7729804515838623, "learning_rate": 3.8e-05, "loss": 0.0355, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9263157894736842, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.923076923076923, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9069767441860465, "eval_ORGANIZATION_recall": 0.9397590361445783, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04491155222058296, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9468479604449938, "eval_overall_precision": 0.9387254901960784, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2895, "eval_samples_per_second": 587.229, "eval_steps_per_second": 10.363, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.5651174783706665, "learning_rate": 3.7500000000000003e-05, "loss": 0.0316, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8725490196078431, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9085365853658537, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9197530864197531, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.046948954463005066, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9318463444857497, "eval_overall_precision": 0.9261083743842364, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2874, "eval_samples_per_second": 591.541, "eval_steps_per_second": 10.439, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.0639706701040268, "learning_rate": 3.7e-05, "loss": 0.0309, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9354838709677419, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9354838709677419, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9337349397590361, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9337349397590361, "eval_ORGANIZATION_recall": 0.9337349397590361, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04571140930056572, "eval_overall_accuracy": 0.9895747599451303, "eval_overall_f1": 0.9464508094645082, "eval_overall_precision": 0.945273631840796, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2884, "eval_samples_per_second": 589.475, "eval_steps_per_second": 10.402, "step": 2496 }, { "epoch": 27.0, "grad_norm": 1.063063383102417, "learning_rate": 3.65e-05, "loss": 0.0293, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9285714285714286, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9176470588235294, "eval_ORGANIZATION_recall": 0.9397590361445783, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04741787165403366, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.942189421894219, "eval_overall_precision": 0.9296116504854369, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2854, "eval_samples_per_second": 595.68, "eval_steps_per_second": 10.512, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.6515504121780396, "learning_rate": 3.6e-05, "loss": 0.0299, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9473684210526316, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9278350515463918, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.920353982300885, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9017341040462428, "eval_ORGANIZATION_recall": 0.9397590361445783, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.050067901611328125, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.944649446494465, "eval_overall_precision": 0.9320388349514563, "eval_overall_recall": 0.9576059850374065, "eval_runtime": 0.2852, "eval_samples_per_second": 596.134, "eval_steps_per_second": 10.52, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.7020676136016846, "learning_rate": 3.55e-05, "loss": 0.0264, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9281437125748503, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9226190476190477, "eval_ORGANIZATION_recall": 0.9337349397590361, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.0471595823764801, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9433497536945813, "eval_overall_precision": 0.9318734793187348, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2873, "eval_samples_per_second": 591.689, "eval_steps_per_second": 10.442, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.7510899901390076, "learning_rate": 3.5e-05, "loss": 0.0257, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.925531914893617, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9157894736842105, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9212827988338192, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8926553672316384, "eval_ORGANIZATION_recall": 0.9518072289156626, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05023224279284477, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9423312883435583, "eval_overall_precision": 0.927536231884058, "eval_overall_recall": 0.9576059850374065, "eval_runtime": 0.2848, "eval_samples_per_second": 596.95, "eval_steps_per_second": 10.534, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.40647125244140625, "learning_rate": 3.45e-05, "loss": 0.0262, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.911917098445596, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.88, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.050489392131567, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9371146732429101, "eval_overall_precision": 0.926829268292683, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2868, "eval_samples_per_second": 592.731, "eval_steps_per_second": 10.46, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.44217225909233093, "learning_rate": 3.4000000000000007e-05, "loss": 0.0263, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9263157894736843, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9072164948453608, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9235474006116209, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.937888198757764, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05121176689863205, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9438202247191011, "eval_overall_precision": 0.945, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2867, "eval_samples_per_second": 593.018, "eval_steps_per_second": 10.465, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.46512940526008606, "learning_rate": 3.35e-05, "loss": 0.0246, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9263157894736842, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.923076923076923, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9069767441860465, "eval_ORGANIZATION_recall": 0.9397590361445783, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04934880882501602, "eval_overall_accuracy": 0.9893004115226337, "eval_overall_f1": 0.9443757725587144, "eval_overall_precision": 0.9362745098039216, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2855, "eval_samples_per_second": 595.425, "eval_steps_per_second": 10.508, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.3086019456386566, "learning_rate": 3.3e-05, "loss": 0.0242, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9161676646706587, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9107142857142857, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04890041425824165, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9407407407407408, "eval_overall_precision": 0.9315403422982885, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2863, "eval_samples_per_second": 593.782, "eval_steps_per_second": 10.479, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.2681832015514374, "learning_rate": 3.2500000000000004e-05, "loss": 0.0217, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.9304812834224598, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.925531914893617, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.911764705882353, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8908045977011494, "eval_ORGANIZATION_recall": 0.9337349397590361, "eval_PERSON_f1": 0.9647887323943662, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9647887323943662, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.05158211290836334, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9346485819975339, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2845, "eval_samples_per_second": 597.526, "eval_steps_per_second": 10.545, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.3382647633552551, "learning_rate": 3.2000000000000005e-05, "loss": 0.0221, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9240121580547112, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.056546807289123535, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9404466501240694, "eval_overall_precision": 0.9358024691358025, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2884, "eval_samples_per_second": 589.516, "eval_steps_per_second": 10.403, "step": 3456 }, { "epoch": 37.0, "grad_norm": 1.3838005065917969, "learning_rate": 3.15e-05, "loss": 0.0219, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.9319371727748691, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04845535010099411, "eval_overall_accuracy": 0.9893004115226337, "eval_overall_f1": 0.9404466501240694, "eval_overall_precision": 0.9358024691358025, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2874, "eval_samples_per_second": 591.486, "eval_steps_per_second": 10.438, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.24594035744667053, "learning_rate": 3.1e-05, "loss": 0.0217, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.918429003021148, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9212121212121213, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9681978798586572, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9716312056737588, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.052869848906993866, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9416149068322982, "eval_overall_precision": 0.9381188118811881, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2858, "eval_samples_per_second": 594.823, "eval_steps_per_second": 10.497, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.40099620819091797, "learning_rate": 3.05e-05, "loss": 0.0236, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.9304812834224598, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.925531914893617, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9107142857142857, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9681978798586572, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9716312056737588, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.05043813958764076, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9354838709677419, "eval_overall_precision": 0.9308641975308642, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2828, "eval_samples_per_second": 601.107, "eval_steps_per_second": 10.608, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.19064545631408691, "learning_rate": 3e-05, "loss": 0.0205, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.90625, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9134328358208955, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9053254437869822, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9614035087719298, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.958041958041958, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.053412336856126785, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9285714285714285, "eval_overall_precision": 0.9172749391727494, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2847, "eval_samples_per_second": 597.038, "eval_steps_per_second": 10.536, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.02482328750193119, "learning_rate": 2.95e-05, "loss": 0.0193, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.9072164948453608, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8712871287128713, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.890207715133531, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8771929824561403, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0644853487610817, "eval_overall_accuracy": 0.9860082304526749, "eval_overall_f1": 0.9249692496924969, "eval_overall_precision": 0.912621359223301, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2865, "eval_samples_per_second": 593.276, "eval_steps_per_second": 10.47, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.5300124287605286, "learning_rate": 2.9e-05, "loss": 0.0198, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.8948948948948949, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8922155688622755, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.0561443567276001, "eval_overall_accuracy": 0.9860082304526749, "eval_overall_f1": 0.9320148331273177, "eval_overall_precision": 0.9240196078431373, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2898, "eval_samples_per_second": 586.544, "eval_steps_per_second": 10.351, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.8966781497001648, "learning_rate": 2.8499999999999998e-05, "loss": 0.018, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9368421052631579, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9175257731958762, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05398855730891228, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9402985074626866, "eval_overall_precision": 0.9379652605459057, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2963, "eval_samples_per_second": 573.712, "eval_steps_per_second": 10.124, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.3985178768634796, "learning_rate": 2.8000000000000003e-05, "loss": 0.0188, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9109947643979057, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8877551020408163, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8975903614457831, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8975903614457831, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05714266002178192, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9293680297397771, "eval_overall_precision": 0.9236453201970444, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2837, "eval_samples_per_second": 599.261, "eval_steps_per_second": 10.575, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.04688119515776634, "learning_rate": 2.7500000000000004e-05, "loss": 0.0177, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.9263157894736843, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9072164948453608, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9030303030303031, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9085365853658537, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05286408215761185, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9353233830845771, "eval_overall_precision": 0.9330024813895782, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2857, "eval_samples_per_second": 595.001, "eval_steps_per_second": 10.5, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.19999141991138458, "learning_rate": 2.7000000000000002e-05, "loss": 0.0178, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.914572864321608, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8584905660377359, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.8895705521472392, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.90625, "eval_ORGANIZATION_recall": 0.8734939759036144, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.053873803466558456, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9282178217821783, "eval_overall_precision": 0.9213759213759214, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2848, "eval_samples_per_second": 597.008, "eval_steps_per_second": 10.535, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.04654427990317345, "learning_rate": 2.6500000000000004e-05, "loss": 0.0168, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.9319371727748691, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9166666666666667, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05449910834431648, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9432098765432098, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.283, "eval_samples_per_second": 600.766, "eval_steps_per_second": 10.602, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.35884276032447815, "learning_rate": 2.6000000000000002e-05, "loss": 0.016, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9096385542168675, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9096385542168675, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05373828485608101, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.287, "eval_samples_per_second": 592.239, "eval_steps_per_second": 10.451, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.010812529362738132, "learning_rate": 2.5500000000000003e-05, "loss": 0.0167, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9319371727748691, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.927710843373494, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.927710843373494, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05074891448020935, "eval_overall_accuracy": 0.9893004115226337, "eval_overall_f1": 0.9478908188585606, "eval_overall_precision": 0.9432098765432099, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.3002, "eval_samples_per_second": 566.205, "eval_steps_per_second": 9.992, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.36007794737815857, "learning_rate": 2.5e-05, "loss": 0.0157, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.9109947643979057, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8877551020408163, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9080118694362017, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8947368421052632, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05863998457789421, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9346485819975339, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2864, "eval_samples_per_second": 593.633, "eval_steps_per_second": 10.476, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.11987640708684921, "learning_rate": 2.45e-05, "loss": 0.015, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.90625, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.052376989275217056, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9344870210135972, "eval_overall_precision": 0.9264705882352942, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2864, "eval_samples_per_second": 593.597, "eval_steps_per_second": 10.475, "step": 4896 }, { "epoch": 52.0, "grad_norm": 0.8939554691314697, "learning_rate": 2.4e-05, "loss": 0.0155, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9025641025641027, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8627450980392157, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9041916167664671, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8988095238095238, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06287980824708939, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9297163995067818, "eval_overall_precision": 0.9195121951219513, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2878, "eval_samples_per_second": 590.774, "eval_steps_per_second": 10.425, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.20487843453884125, "learning_rate": 2.35e-05, "loss": 0.0156, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9015544041450777, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.87, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9216867469879518, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9216867469879518, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06014350429177284, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9356435643564357, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2877, "eval_samples_per_second": 590.819, "eval_steps_per_second": 10.426, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.055074214935302734, "learning_rate": 2.3000000000000003e-05, "loss": 0.0141, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.9081632653061223, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8640776699029126, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9085545722713864, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8901734104046243, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.07066118717193604, "eval_overall_accuracy": 0.9849108367626886, "eval_overall_f1": 0.9326805385556914, "eval_overall_precision": 0.9158653846153846, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.286, "eval_samples_per_second": 594.39, "eval_steps_per_second": 10.489, "step": 5184 }, { "epoch": 55.0, "grad_norm": 0.6687442064285278, "learning_rate": 2.25e-05, "loss": 0.0133, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9146341463414634, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9259259259259259, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06329189985990524, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9378109452736318, "eval_overall_precision": 0.9354838709677419, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2846, "eval_samples_per_second": 597.26, "eval_steps_per_second": 10.54, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.35132700204849243, "learning_rate": 2.2000000000000003e-05, "loss": 0.013, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9216867469879518, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9216867469879518, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9681978798586572, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9716312056737588, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.059241946786642075, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9379652605459058, "eval_overall_precision": 0.9333333333333333, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2858, "eval_samples_per_second": 594.775, "eval_steps_per_second": 10.496, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.6489832401275635, "learning_rate": 2.15e-05, "loss": 0.0143, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8725490196078431, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9207317073170733, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9320987654320988, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.0567299947142601, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2856, "eval_samples_per_second": 595.273, "eval_steps_per_second": 10.505, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.1813511699438095, "learning_rate": 2.1e-05, "loss": 0.0135, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9240121580547112, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.06323638558387756, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9429280397022332, "eval_overall_precision": 0.9382716049382716, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2848, "eval_samples_per_second": 596.907, "eval_steps_per_second": 10.534, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.027425482869148254, "learning_rate": 2.05e-05, "loss": 0.0111, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9263157894736843, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9072164948453608, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9174311926605505, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9316770186335404, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9681978798586572, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9716312056737588, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.06284485757350922, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9375, "eval_overall_precision": 0.9398496240601504, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2875, "eval_samples_per_second": 591.307, "eval_steps_per_second": 10.435, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.3114117383956909, "learning_rate": 2e-05, "loss": 0.0108, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.9157894736842105, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8969072164948454, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9681978798586572, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9716312056737588, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.05974132940173149, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9330024813895781, "eval_overall_precision": 0.928395061728395, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2878, "eval_samples_per_second": 590.651, "eval_steps_per_second": 10.423, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.14523079991340637, "learning_rate": 1.9500000000000003e-05, "loss": 0.0144, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9169230769230768, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9371069182389937, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.9681978798586572, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9716312056737588, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.06437654048204422, "eval_overall_accuracy": 0.9860082304526749, "eval_overall_f1": 0.935, "eval_overall_precision": 0.9373433583959899, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2889, "eval_samples_per_second": 588.359, "eval_steps_per_second": 10.383, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.9911012649536133, "learning_rate": 1.9e-05, "loss": 0.0124, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.9109947643979057, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8877551020408163, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9166666666666667, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05878928676247597, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9369592088998765, "eval_overall_precision": 0.928921568627451, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2873, "eval_samples_per_second": 591.665, "eval_steps_per_second": 10.441, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.3812105059623718, "learning_rate": 1.85e-05, "loss": 0.0117, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9281437125748503, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9226190476190477, "eval_ORGANIZATION_recall": 0.9337349397590361, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0626605674624443, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9442379182156134, "eval_overall_precision": 0.9384236453201971, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2847, "eval_samples_per_second": 597.174, "eval_steps_per_second": 10.538, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.28910690546035767, "learning_rate": 1.8e-05, "loss": 0.0101, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.9263157894736843, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9072164948453608, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.924924924924925, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9221556886227545, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0623946376144886, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9440993788819876, "eval_overall_precision": 0.9405940594059405, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2842, "eval_samples_per_second": 598.224, "eval_steps_per_second": 10.557, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.6301209926605225, "learning_rate": 1.75e-05, "loss": 0.0117, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9047619047619048, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06263893097639084, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9356435643564357, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2845, "eval_samples_per_second": 597.536, "eval_steps_per_second": 10.545, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.4729289412498474, "learning_rate": 1.7000000000000003e-05, "loss": 0.0108, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9052631578947367, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8865979381443299, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.9179331306990882, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9647887323943662, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9647887323943662, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.05901220068335533, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9315068493150686, "eval_overall_precision": 0.9303482587064676, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2843, "eval_samples_per_second": 597.975, "eval_steps_per_second": 10.553, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.20351357758045197, "learning_rate": 1.65e-05, "loss": 0.0104, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9161676646706587, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9107142857142857, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06248219311237335, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9405940594059405, "eval_overall_precision": 0.9336609336609336, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2859, "eval_samples_per_second": 594.682, "eval_steps_per_second": 10.494, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.7123875617980957, "learning_rate": 1.6000000000000003e-05, "loss": 0.0119, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9319371727748691, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9166666666666667, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9058823529411765, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06472578644752502, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2854, "eval_samples_per_second": 595.638, "eval_steps_per_second": 10.511, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.5229490399360657, "learning_rate": 1.55e-05, "loss": 0.011, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05594140291213989, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9379652605459058, "eval_overall_precision": 0.9333333333333333, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2862, "eval_samples_per_second": 594.072, "eval_steps_per_second": 10.484, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.36743980646133423, "learning_rate": 1.5e-05, "loss": 0.0096, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05785645917057991, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2866, "eval_samples_per_second": 593.127, "eval_steps_per_second": 10.467, "step": 6720 }, { "epoch": 71.0, "grad_norm": 1.1272770166397095, "learning_rate": 1.45e-05, "loss": 0.0103, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9047619047619048, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8941176470588236, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06020239740610123, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9344870210135972, "eval_overall_precision": 0.9264705882352942, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2831, "eval_samples_per_second": 600.523, "eval_steps_per_second": 10.597, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.5486194491386414, "learning_rate": 1.4000000000000001e-05, "loss": 0.0091, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9212121212121213, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.926829268292683, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0652991458773613, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9416149068322982, "eval_overall_precision": 0.9381188118811881, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2872, "eval_samples_per_second": 591.962, "eval_steps_per_second": 10.446, "step": 6912 }, { "epoch": 73.0, "grad_norm": 1.128568172454834, "learning_rate": 1.3500000000000001e-05, "loss": 0.0093, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9096385542168675, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9096385542168675, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9681978798586572, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9716312056737588, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.06982634961605072, "eval_overall_accuracy": 0.9865569272976681, "eval_overall_f1": 0.9318463444857497, "eval_overall_precision": 0.9261083743842364, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2848, "eval_samples_per_second": 596.928, "eval_steps_per_second": 10.534, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.033605970442295074, "learning_rate": 1.3000000000000001e-05, "loss": 0.0091, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9263157894736843, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9072164948453608, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9074626865671641, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8994082840236687, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06609321385622025, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9356435643564357, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2875, "eval_samples_per_second": 591.305, "eval_steps_per_second": 10.435, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.5324849486351013, "learning_rate": 1.25e-05, "loss": 0.0093, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.90625, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9009009009009009, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8982035928143712, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9681978798586572, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9716312056737588, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.07267607003450394, "eval_overall_accuracy": 0.9854595336076818, "eval_overall_f1": 0.9257425742574258, "eval_overall_precision": 0.918918918918919, "eval_overall_recall": 0.9326683291770573, "eval_runtime": 0.2867, "eval_samples_per_second": 592.871, "eval_steps_per_second": 10.462, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.02331961691379547, "learning_rate": 1.2e-05, "loss": 0.0087, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06968007981777191, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9369592088998765, "eval_overall_precision": 0.928921568627451, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2872, "eval_samples_per_second": 591.918, "eval_steps_per_second": 10.446, "step": 7296 }, { "epoch": 77.0, "grad_norm": 1.1283886432647705, "learning_rate": 1.1500000000000002e-05, "loss": 0.0104, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.911917098445596, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.88, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9212121212121213, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.926829268292683, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06767893582582474, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9391304347826087, "eval_overall_precision": 0.9356435643564357, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2844, "eval_samples_per_second": 597.853, "eval_steps_per_second": 10.55, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.6260883212089539, "learning_rate": 1.1000000000000001e-05, "loss": 0.0089, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9109947643979057, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8877551020408163, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.906906906906907, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9041916167664671, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9681978798586572, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9716312056737588, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.06810528039932251, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9293680297397771, "eval_overall_precision": 0.9236453201970444, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2859, "eval_samples_per_second": 594.623, "eval_steps_per_second": 10.493, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.1493467390537262, "learning_rate": 1.05e-05, "loss": 0.0092, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9014925373134328, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.893491124260355, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06890777498483658, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9320148331273177, "eval_overall_precision": 0.9240196078431373, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2852, "eval_samples_per_second": 596.158, "eval_steps_per_second": 10.52, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.6644344329833984, "learning_rate": 1e-05, "loss": 0.0069, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9047619047619048, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.07149485498666763, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9356435643564357, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2828, "eval_samples_per_second": 601.1, "eval_steps_per_second": 10.608, "step": 7680 }, { "epoch": 81.0, "grad_norm": 1.0200001001358032, "learning_rate": 9.5e-06, "loss": 0.0087, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9074626865671641, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8994082840236687, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06909173727035522, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9344870210135972, "eval_overall_precision": 0.9264705882352942, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2859, "eval_samples_per_second": 594.513, "eval_steps_per_second": 10.491, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.21058472990989685, "learning_rate": 9e-06, "loss": 0.0081, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.906906906906907, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9041916167664671, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.07346898317337036, "eval_overall_accuracy": 0.9860082304526749, "eval_overall_f1": 0.9343246592317225, "eval_overall_precision": 0.9285714285714286, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2865, "eval_samples_per_second": 593.316, "eval_steps_per_second": 10.47, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.9036238789558411, "learning_rate": 8.500000000000002e-06, "loss": 0.0099, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06314818561077118, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2854, "eval_samples_per_second": 595.652, "eval_steps_per_second": 10.512, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.02120956964790821, "learning_rate": 8.000000000000001e-06, "loss": 0.0085, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.918429003021148, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9212121212121213, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06590625643730164, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9391304347826087, "eval_overall_precision": 0.9356435643564357, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2843, "eval_samples_per_second": 597.939, "eval_steps_per_second": 10.552, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.7899487018585205, "learning_rate": 7.5e-06, "loss": 0.0092, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9216867469879518, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9216867469879518, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06632930040359497, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9405940594059405, "eval_overall_precision": 0.9336609336609336, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2872, "eval_samples_per_second": 591.847, "eval_steps_per_second": 10.444, "step": 8160 }, { "epoch": 86.0, "grad_norm": 1.0977509021759033, "learning_rate": 7.000000000000001e-06, "loss": 0.0076, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.906906906906907, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9041916167664671, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06624973565340042, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9343246592317225, "eval_overall_precision": 0.9285714285714286, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2852, "eval_samples_per_second": 596.122, "eval_steps_per_second": 10.52, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.06188450753688812, "learning_rate": 6.5000000000000004e-06, "loss": 0.0073, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.918429003021148, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9212121212121213, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06855077296495438, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9429280397022332, "eval_overall_precision": 0.9382716049382716, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2882, "eval_samples_per_second": 589.847, "eval_steps_per_second": 10.409, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.21970857679843903, "learning_rate": 6e-06, "loss": 0.0068, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06893233209848404, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9366459627329192, "eval_overall_precision": 0.9331683168316832, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2865, "eval_samples_per_second": 593.348, "eval_steps_per_second": 10.471, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.05854140594601631, "learning_rate": 5.500000000000001e-06, "loss": 0.0076, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06832458078861237, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9356435643564357, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2851, "eval_samples_per_second": 596.36, "eval_steps_per_second": 10.524, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.20013177394866943, "learning_rate": 5e-06, "loss": 0.0071, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06415162235498428, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9417596034696407, "eval_overall_precision": 0.9359605911330049, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2848, "eval_samples_per_second": 596.83, "eval_steps_per_second": 10.532, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.0871826782822609, "learning_rate": 4.5e-06, "loss": 0.0069, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9047619047619048, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.07019613683223724, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9356435643564357, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2853, "eval_samples_per_second": 595.914, "eval_steps_per_second": 10.516, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.16851916909217834, "learning_rate": 4.000000000000001e-06, "loss": 0.0073, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06724745780229568, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9368029739776952, "eval_overall_precision": 0.9310344827586207, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2877, "eval_samples_per_second": 590.936, "eval_steps_per_second": 10.428, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.19750891625881195, "learning_rate": 3.5000000000000004e-06, "loss": 0.007, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06532972306013107, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9368029739776952, "eval_overall_precision": 0.9310344827586207, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2934, "eval_samples_per_second": 579.374, "eval_steps_per_second": 10.224, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.07180120795965195, "learning_rate": 3e-06, "loss": 0.007, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.06773727387189865, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2941, "eval_samples_per_second": 578.014, "eval_steps_per_second": 10.2, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.31866025924682617, "learning_rate": 2.5e-06, "loss": 0.0077, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.0693470910191536, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9344870210135972, "eval_overall_precision": 0.9264705882352942, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2917, "eval_samples_per_second": 582.839, "eval_steps_per_second": 10.285, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.8485398292541504, "learning_rate": 2.0000000000000003e-06, "loss": 0.0071, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9047619047619048, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.07040981203317642, "eval_overall_accuracy": 0.9871056241426612, "eval_overall_f1": 0.9333333333333333, "eval_overall_precision": 0.9242053789731052, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2967, "eval_samples_per_second": 572.951, "eval_steps_per_second": 10.111, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.03155836462974548, "learning_rate": 1.5e-06, "loss": 0.007, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.069269560277462, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9356435643564357, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2925, "eval_samples_per_second": 581.238, "eval_steps_per_second": 10.257, "step": 9312 }, { "epoch": 98.0, "grad_norm": 0.1880914568901062, "learning_rate": 1.0000000000000002e-06, "loss": 0.0062, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06927915662527084, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9368029739776952, "eval_overall_precision": 0.9310344827586207, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2925, "eval_samples_per_second": 581.127, "eval_steps_per_second": 10.255, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.29024115204811096, "learning_rate": 5.000000000000001e-07, "loss": 0.0062, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06901175528764725, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9368029739776952, "eval_overall_precision": 0.9310344827586207, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2921, "eval_samples_per_second": 582.043, "eval_steps_per_second": 10.271, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.5472394824028015, "learning_rate": 0.0, "loss": 0.0068, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9166666666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06890885531902313, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9368029739776952, "eval_overall_precision": 0.9310344827586207, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2907, "eval_samples_per_second": 584.734, "eval_steps_per_second": 10.319, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 4208466347875278.0, "train_loss": 0.03755435147012273, "train_runtime": 532.4294, "train_samples_per_second": 287.55, "train_steps_per_second": 18.031 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 4208466347875278.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }