{ "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": 1.0768136978149414, "learning_rate": 4.9500000000000004e-05, "loss": 0.8532, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.0, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.0, "eval_LOCATION_recall": 0.0, "eval_ORGANIZATION_f1": 0.22784810126582278, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.24, "eval_ORGANIZATION_recall": 0.21686746987951808, "eval_PERSON_f1": 0.28104575163398693, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.2621951219512195, "eval_PERSON_recall": 0.3028169014084507, "eval_loss": 0.3947194218635559, "eval_overall_accuracy": 0.863923182441701, "eval_overall_f1": 0.2203626220362622, "eval_overall_precision": 0.25, "eval_overall_recall": 0.1970074812967581, "eval_runtime": 0.2958, "eval_samples_per_second": 574.685, "eval_steps_per_second": 10.142, "step": 96 }, { "epoch": 2.0, "grad_norm": 0.9138985276222229, "learning_rate": 4.9e-05, "loss": 0.359, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.3900414937759336, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.31756756756756754, "eval_LOCATION_recall": 0.5053763440860215, "eval_ORGANIZATION_f1": 0.5906432748538012, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.5738636363636364, "eval_ORGANIZATION_recall": 0.608433734939759, "eval_PERSON_f1": 0.7033639143730888, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.6216216216216216, "eval_PERSON_recall": 0.8098591549295775, "eval_loss": 0.22303983569145203, "eval_overall_accuracy": 0.934156378600823, "eval_overall_f1": 0.5780219780219781, "eval_overall_precision": 0.5166994106090373, "eval_overall_recall": 0.655860349127182, "eval_runtime": 0.2946, "eval_samples_per_second": 577.114, "eval_steps_per_second": 10.184, "step": 192 }, { "epoch": 3.0, "grad_norm": 0.8776483535766602, "learning_rate": 4.85e-05, "loss": 0.1925, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.726190476190476, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8133333333333334, "eval_LOCATION_recall": 0.6559139784946236, "eval_ORGANIZATION_f1": 0.7331536388140161, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.6634146341463415, "eval_ORGANIZATION_recall": 0.8192771084337349, "eval_PERSON_f1": 0.9355932203389831, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9019607843137255, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.11146330088376999, "eval_overall_accuracy": 0.9613168724279836, "eval_overall_f1": 0.803357314148681, "eval_overall_precision": 0.7736720554272517, "eval_overall_recall": 0.8354114713216958, "eval_runtime": 0.2924, "eval_samples_per_second": 581.304, "eval_steps_per_second": 10.258, "step": 288 }, { "epoch": 4.0, "grad_norm": 1.1688709259033203, "learning_rate": 4.8e-05, "loss": 0.1309, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.8018867924528302, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7142857142857143, "eval_LOCATION_recall": 0.9139784946236559, "eval_ORGANIZATION_f1": 0.8071216617210683, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.7953216374269005, "eval_ORGANIZATION_recall": 0.8192771084337349, "eval_PERSON_f1": 0.9556313993174063, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9271523178807947, "eval_PERSON_recall": 0.9859154929577465, "eval_loss": 0.08377877622842789, "eval_overall_accuracy": 0.9720164609053498, "eval_overall_f1": 0.857482185273159, "eval_overall_precision": 0.81859410430839, "eval_overall_recall": 0.9002493765586035, "eval_runtime": 0.2931, "eval_samples_per_second": 580.016, "eval_steps_per_second": 10.236, "step": 384 }, { "epoch": 5.0, "grad_norm": 0.4364277422428131, "learning_rate": 4.75e-05, "loss": 0.1058, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.8756756756756756, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8804347826086957, "eval_LOCATION_recall": 0.8709677419354839, "eval_ORGANIZATION_f1": 0.8742514970059881, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8690476190476191, "eval_ORGANIZATION_recall": 0.8795180722891566, "eval_PERSON_f1": 0.972027972027972, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9652777777777778, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.06439696997404099, "eval_overall_accuracy": 0.9810699588477366, "eval_overall_f1": 0.9093167701863354, "eval_overall_precision": 0.905940594059406, "eval_overall_recall": 0.912718204488778, "eval_runtime": 0.2956, "eval_samples_per_second": 575.164, "eval_steps_per_second": 10.15, "step": 480 }, { "epoch": 6.0, "grad_norm": 0.9933373332023621, "learning_rate": 4.7e-05, "loss": 0.0954, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.8502415458937198, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7719298245614035, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.853582554517134, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8838709677419355, "eval_ORGANIZATION_recall": 0.8253012048192772, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05711955949664116, "eval_overall_accuracy": 0.9818930041152263, "eval_overall_f1": 0.8940886699507389, "eval_overall_precision": 0.8832116788321168, "eval_overall_recall": 0.9052369077306733, "eval_runtime": 0.2951, "eval_samples_per_second": 576.101, "eval_steps_per_second": 10.166, "step": 576 }, { "epoch": 7.0, "grad_norm": 1.0747640132904053, "learning_rate": 4.6500000000000005e-05, "loss": 0.0835, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.8775510204081631, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8349514563106796, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.9043478260869566, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8715083798882681, "eval_ORGANIZATION_recall": 0.9397590361445783, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05005069077014923, "eval_overall_accuracy": 0.984636488340192, "eval_overall_f1": 0.9200968523002423, "eval_overall_precision": 0.8941176470588236, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2947, "eval_samples_per_second": 576.819, "eval_steps_per_second": 10.179, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.2273331880569458, "learning_rate": 4.600000000000001e-05, "loss": 0.0757, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.8855721393034827, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8240740740740741, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9003021148036254, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9030303030303031, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.051483578979969025, "eval_overall_accuracy": 0.9832647462277092, "eval_overall_f1": 0.9238329238329237, "eval_overall_precision": 0.910411622276029, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2936, "eval_samples_per_second": 579.097, "eval_steps_per_second": 10.219, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.4562922716140747, "learning_rate": 4.55e-05, "loss": 0.0712, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.883248730964467, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8365384615384616, "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": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04515055567026138, "eval_overall_accuracy": 0.9851851851851852, "eval_overall_f1": 0.9262899262899262, "eval_overall_precision": 0.9128329297820823, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2921, "eval_samples_per_second": 582.059, "eval_steps_per_second": 10.272, "step": 864 }, { "epoch": 10.0, "grad_norm": 1.2254160642623901, "learning_rate": 4.5e-05, "loss": 0.0653, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.8613861386138614, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.7981651376146789, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.8990825688073394, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9130434782608695, "eval_ORGANIZATION_recall": 0.8855421686746988, "eval_PERSON_f1": 0.972027972027972, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9652777777777778, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.042413756251335144, "eval_overall_accuracy": 0.9838134430727024, "eval_overall_f1": 0.9153374233128836, "eval_overall_precision": 0.9009661835748792, "eval_overall_recall": 0.9301745635910225, "eval_runtime": 0.2946, "eval_samples_per_second": 576.963, "eval_steps_per_second": 10.182, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.46680667996406555, "learning_rate": 4.4500000000000004e-05, "loss": 0.0602, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.883248730964467, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8365384615384616, "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": 0.9686411149825783, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9586206896551724, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04325629398226738, "eval_overall_accuracy": 0.9854595336076818, "eval_overall_f1": 0.9253365973072215, "eval_overall_precision": 0.9086538461538461, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2959, "eval_samples_per_second": 574.541, "eval_steps_per_second": 10.139, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.5539126992225647, "learning_rate": 4.4000000000000006e-05, "loss": 0.0517, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.8969072164948454, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8613861386138614, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.913946587537092, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9005847953216374, "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": 0.03981715440750122, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.932349323493235, "eval_overall_precision": 0.9199029126213593, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2928, "eval_samples_per_second": 580.593, "eval_steps_per_second": 10.246, "step": 1152 }, { "epoch": 13.0, "grad_norm": 0.7579489350318909, "learning_rate": 4.35e-05, "loss": 0.0504, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.9297297297297297, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9347826086956522, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.9047619047619048, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.8941176470588236, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9547038327526133, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9448275862068966, "eval_PERSON_recall": 0.9647887323943662, "eval_loss": 0.04784214869141579, "eval_overall_accuracy": 0.9854595336076818, "eval_overall_f1": 0.9282178217821783, "eval_overall_precision": 0.9213759213759214, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2932, "eval_samples_per_second": 579.856, "eval_steps_per_second": 10.233, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.31520748138427734, "learning_rate": 4.3e-05, "loss": 0.0475, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.9035532994923857, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8557692307692307, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9292307692307692, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.949685534591195, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.9686411149825783, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9586206896551724, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04845843464136124, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9369592088998765, "eval_overall_precision": 0.928921568627451, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2926, "eval_samples_per_second": 580.905, "eval_steps_per_second": 10.251, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.580851137638092, "learning_rate": 4.25e-05, "loss": 0.0467, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.9000000000000001, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8411214953271028, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9012345679012346, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9240506329113924, "eval_ORGANIZATION_recall": 0.8795180722891566, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04513963684439659, "eval_overall_accuracy": 0.9857338820301783, "eval_overall_f1": 0.9270704573547589, "eval_overall_precision": 0.9191176470588235, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2927, "eval_samples_per_second": 580.788, "eval_steps_per_second": 10.249, "step": 1440 }, { "epoch": 16.0, "grad_norm": 0.8288644552230835, "learning_rate": 4.2e-05, "loss": 0.044, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.911917098445596, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.88, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9090909090909091, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9146341463414634, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04263924062252045, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9318463444857497, "eval_overall_precision": 0.9261083743842364, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.294, "eval_samples_per_second": 578.226, "eval_steps_per_second": 10.204, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.507310152053833, "learning_rate": 4.15e-05, "loss": 0.0409, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9123867069486404, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9151515151515152, "eval_ORGANIZATION_recall": 0.9096385542168675, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04281206056475639, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9346485819975339, "eval_overall_precision": 0.9243902439024391, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2926, "eval_samples_per_second": 580.917, "eval_steps_per_second": 10.251, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.47665178775787354, "learning_rate": 4.1e-05, "loss": 0.037, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.9278350515463919, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9135802469135801, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9367088607594937, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04476248100399971, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9376558603491272, "eval_overall_precision": 0.9376558603491272, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2945, "eval_samples_per_second": 577.326, "eval_steps_per_second": 10.188, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.8693915009498596, "learning_rate": 4.05e-05, "loss": 0.0357, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.9574468085106383, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9473684210526315, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9300911854103343, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9386503067484663, "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": 0.045135971158742905, "eval_overall_accuracy": 0.9895747599451303, "eval_overall_f1": 0.9524999999999999, "eval_overall_precision": 0.9548872180451128, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2934, "eval_samples_per_second": 579.421, "eval_steps_per_second": 10.225, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.47152310609817505, "learning_rate": 4e-05, "loss": 0.032, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.911917098445596, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.88, "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": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04165427014231682, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9407407407407408, "eval_overall_precision": 0.9315403422982885, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2935, "eval_samples_per_second": 579.202, "eval_steps_per_second": 10.221, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.17908351123332977, "learning_rate": 3.9500000000000005e-05, "loss": 0.0326, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "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": 0.04387632757425308, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9394313967861558, "eval_overall_precision": 0.9313725490196079, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2936, "eval_samples_per_second": 579.049, "eval_steps_per_second": 10.219, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.20579631626605988, "learning_rate": 3.9000000000000006e-05, "loss": 0.0348, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.9333333333333333, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8921568627450981, "eval_LOCATION_recall": 0.978494623655914, "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": 0.04139135405421257, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9395807644882862, "eval_overall_precision": 0.9292682926829269, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2938, "eval_samples_per_second": 578.717, "eval_steps_per_second": 10.213, "step": 2112 }, { "epoch": 23.0, "grad_norm": 0.6211321353912354, "learning_rate": 3.85e-05, "loss": 0.0288, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.9479166666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9191919191919192, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9258160237388725, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9122807017543859, "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": 0.04008758068084717, "eval_overall_accuracy": 0.9901234567901235, "eval_overall_f1": 0.949569495694957, "eval_overall_precision": 0.9368932038834952, "eval_overall_recall": 0.9625935162094763, "eval_runtime": 0.2952, "eval_samples_per_second": 575.839, "eval_steps_per_second": 10.162, "step": 2208 }, { "epoch": 24.0, "grad_norm": 0.758486270904541, "learning_rate": 3.8e-05, "loss": 0.0317, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9417989417989417, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9270833333333334, "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": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04227767139673233, "eval_overall_accuracy": 0.9895747599451303, "eval_overall_f1": 0.9478908188585606, "eval_overall_precision": 0.9432098765432099, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2922, "eval_samples_per_second": 581.848, "eval_steps_per_second": 10.268, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.6690868139266968, "learning_rate": 3.7500000000000003e-05, "loss": 0.0296, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.9528795811518325, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9285714285714286, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9333333333333333, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9390243902439024, "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": 0.04354260861873627, "eval_overall_accuracy": 0.9901234567901235, "eval_overall_f1": 0.9515527950310558, "eval_overall_precision": 0.948019801980198, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2922, "eval_samples_per_second": 581.89, "eval_steps_per_second": 10.269, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.1914530098438263, "learning_rate": 3.7e-05, "loss": 0.0269, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9300911854103343, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9386503067484663, "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": 0.04295233637094498, "eval_overall_accuracy": 0.9895747599451303, "eval_overall_f1": 0.9477611940298507, "eval_overall_precision": 0.9454094292803971, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2938, "eval_samples_per_second": 578.634, "eval_steps_per_second": 10.211, "step": 2496 }, { "epoch": 27.0, "grad_norm": 0.4813244342803955, "learning_rate": 3.65e-05, "loss": 0.0248, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9473684210526316, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9278350515463918, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.924924924924925, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9221556886227545, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04276817664504051, "eval_overall_accuracy": 0.9898491083676269, "eval_overall_f1": 0.9478908188585606, "eval_overall_precision": 0.9432098765432099, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2917, "eval_samples_per_second": 582.867, "eval_steps_per_second": 10.286, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.7333148121833801, "learning_rate": 3.6e-05, "loss": 0.0263, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9258160237388725, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9122807017543859, "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": 0.04888397827744484, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9408866995073891, "eval_overall_precision": 0.9294403892944039, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2935, "eval_samples_per_second": 579.298, "eval_steps_per_second": 10.223, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.5005503296852112, "learning_rate": 3.55e-05, "loss": 0.0245, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9381443298969072, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.900990099009901, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.911854103343465, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9202453987730062, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04651840776205063, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2966, "eval_samples_per_second": 573.248, "eval_steps_per_second": 10.116, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.17724396288394928, "learning_rate": 3.5e-05, "loss": 0.0217, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.9278350515463919, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9253731343283581, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9171597633136095, "eval_ORGANIZATION_recall": 0.9337349397590361, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.04648192971944809, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.942189421894219, "eval_overall_precision": 0.9296116504854369, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2929, "eval_samples_per_second": 580.308, "eval_steps_per_second": 10.241, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.10259954631328583, "learning_rate": 3.45e-05, "loss": 0.0219, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04794708639383316, "eval_overall_accuracy": 0.9893004115226337, "eval_overall_f1": 0.9443757725587144, "eval_overall_precision": 0.9362745098039216, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2935, "eval_samples_per_second": 579.15, "eval_steps_per_second": 10.22, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.8183249831199646, "learning_rate": 3.4000000000000007e-05, "loss": 0.0209, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9381443298969072, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.900990099009901, "eval_LOCATION_recall": 0.978494623655914, "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": 0.05001248046755791, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9442379182156134, "eval_overall_precision": 0.9384236453201971, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2945, "eval_samples_per_second": 577.309, "eval_steps_per_second": 10.188, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.1250787377357483, "learning_rate": 3.35e-05, "loss": 0.0206, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.9528795811518325, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9285714285714286, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9216867469879518, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9216867469879518, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.04647345840930939, "eval_overall_accuracy": 0.9893004115226337, "eval_overall_f1": 0.949194547707559, "eval_overall_precision": 0.9433497536945813, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2947, "eval_samples_per_second": 576.801, "eval_steps_per_second": 10.179, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.8755994439125061, "learning_rate": 3.3e-05, "loss": 0.0202, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.9473684210526316, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9278350515463918, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9361702127659575, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9447852760736196, "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": 0.0446864478290081, "eval_overall_accuracy": 0.9901234567901235, "eval_overall_f1": 0.9514321295143213, "eval_overall_precision": 0.9502487562189055, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.297, "eval_samples_per_second": 572.432, "eval_steps_per_second": 10.102, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.22532738745212555, "learning_rate": 3.2500000000000004e-05, "loss": 0.0187, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.9347826086956522, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.945054945054945, "eval_LOCATION_recall": 0.9247311827956989, "eval_ORGANIZATION_f1": 0.9266862170087975, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9028571428571428, "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": 0.0517609640955925, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9432098765432098, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2949, "eval_samples_per_second": 576.442, "eval_steps_per_second": 10.173, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.44033437967300415, "learning_rate": 3.2000000000000005e-05, "loss": 0.019, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9212121212121213, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.926829268292683, "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": 0.04985504224896431, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9417596034696407, "eval_overall_precision": 0.9359605911330049, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2936, "eval_samples_per_second": 579.066, "eval_steps_per_second": 10.219, "step": 3456 }, { "epoch": 37.0, "grad_norm": 0.25371018052101135, "learning_rate": 3.15e-05, "loss": 0.0197, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.9456521739130435, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9560439560439561, "eval_LOCATION_recall": 0.9354838709677419, "eval_ORGANIZATION_f1": 0.9432835820895522, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9349112426035503, "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": 0.04717753455042839, "eval_overall_accuracy": 0.9895747599451303, "eval_overall_f1": 0.9527363184079602, "eval_overall_precision": 0.9503722084367245, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2935, "eval_samples_per_second": 579.147, "eval_steps_per_second": 10.22, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.22911065816879272, "learning_rate": 3.1e-05, "loss": 0.0169, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8725490196078431, "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": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05635219067335129, "eval_overall_accuracy": 0.9860082304526749, "eval_overall_f1": 0.9359605911330049, "eval_overall_precision": 0.9245742092457421, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2953, "eval_samples_per_second": 575.636, "eval_steps_per_second": 10.158, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.6669800877571106, "learning_rate": 3.05e-05, "loss": 0.0176, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.911854103343465, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9202453987730062, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05298105254769325, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9369592088998765, "eval_overall_precision": 0.928921568627451, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2936, "eval_samples_per_second": 579.045, "eval_steps_per_second": 10.218, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.30945512652397156, "learning_rate": 3e-05, "loss": 0.0184, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.956989247311828, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.956989247311828, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9285714285714286, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9176470588235294, "eval_ORGANIZATION_recall": 0.9397590361445783, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.049056798219680786, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.949194547707559, "eval_overall_precision": 0.9433497536945813, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.293, "eval_samples_per_second": 580.131, "eval_steps_per_second": 10.238, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.12278193980455399, "learning_rate": 2.95e-05, "loss": 0.0169, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.946236559139785, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.946236559139785, "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": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05431058630347252, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.946716232961586, "eval_overall_precision": 0.9408866995073891, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2914, "eval_samples_per_second": 583.341, "eval_steps_per_second": 10.294, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.2668594419956207, "learning_rate": 2.9e-05, "loss": 0.0152, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9365558912386707, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9393939393939394, "eval_ORGANIZATION_recall": 0.9337349397590361, "eval_PERSON_f1": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.05379322171211243, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9516728624535316, "eval_overall_precision": 0.9458128078817734, "eval_overall_recall": 0.9576059850374065, "eval_runtime": 0.2922, "eval_samples_per_second": 581.788, "eval_steps_per_second": 10.267, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.11598949134349823, "learning_rate": 2.8499999999999998e-05, "loss": 0.015, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9417989417989417, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9270833333333334, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9369369369369371, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9341317365269461, "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": 0.053120292723178864, "eval_overall_accuracy": 0.9898491083676269, "eval_overall_f1": 0.9516728624535316, "eval_overall_precision": 0.9458128078817734, "eval_overall_recall": 0.9576059850374065, "eval_runtime": 0.2934, "eval_samples_per_second": 579.382, "eval_steps_per_second": 10.224, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.2693626582622528, "learning_rate": 2.8000000000000003e-05, "loss": 0.0154, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9368421052631579, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9175257731958762, "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": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05610647052526474, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9443757725587144, "eval_overall_precision": 0.9362745098039216, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2951, "eval_samples_per_second": 576.031, "eval_steps_per_second": 10.165, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.11329390108585358, "learning_rate": 2.7500000000000004e-05, "loss": 0.0149, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.9319371727748691, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.918429003021148, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9212121212121213, "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": 0.05618666484951973, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9416149068322982, "eval_overall_precision": 0.9381188118811881, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.3012, "eval_samples_per_second": 564.356, "eval_steps_per_second": 9.959, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.32401782274246216, "learning_rate": 2.7000000000000002e-05, "loss": 0.0158, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.9374999999999999, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.967741935483871, "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": 0.048293400555849075, "eval_overall_accuracy": 0.9898491083676269, "eval_overall_f1": 0.949194547707559, "eval_overall_precision": 0.9433497536945813, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2913, "eval_samples_per_second": 583.503, "eval_steps_per_second": 10.297, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.02720675617456436, "learning_rate": 2.6500000000000004e-05, "loss": 0.0131, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.9417989417989417, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9270833333333334, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9194029850746269, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9112426035502958, "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": 0.04851473495364189, "eval_overall_accuracy": 0.9893004115226337, "eval_overall_f1": 0.9430693069306931, "eval_overall_precision": 0.9361179361179361, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2927, "eval_samples_per_second": 580.849, "eval_steps_per_second": 10.25, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.11735498160123825, "learning_rate": 2.6000000000000002e-05, "loss": 0.0149, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.91131498470948, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9254658385093167, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.9787234042553192, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9857142857142858, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.059497687965631485, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9376558603491272, "eval_overall_precision": 0.9376558603491272, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2963, "eval_samples_per_second": 573.804, "eval_steps_per_second": 10.126, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.13654810190200806, "learning_rate": 2.5500000000000003e-05, "loss": 0.0147, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8737864077669902, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9051987767584098, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9192546583850931, "eval_ORGANIZATION_recall": 0.891566265060241, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05872794985771179, "eval_overall_accuracy": 0.9862825788751715, "eval_overall_f1": 0.9318463444857497, "eval_overall_precision": 0.9261083743842364, "eval_overall_recall": 0.9376558603491272, "eval_runtime": 0.2942, "eval_samples_per_second": 577.811, "eval_steps_per_second": 10.197, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.09901689738035202, "learning_rate": 2.5e-05, "loss": 0.0128, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.9473684210526316, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9278350515463918, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9194029850746269, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9112426035502958, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05709797516465187, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9455445544554455, "eval_overall_precision": 0.9385749385749386, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2924, "eval_samples_per_second": 581.359, "eval_steps_per_second": 10.259, "step": 4800 }, { "epoch": 51.0, "grad_norm": 1.0321968793869019, "learning_rate": 2.45e-05, "loss": 0.0107, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.918429003021148, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9212121212121213, "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": 0.06082933023571968, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9381188118811881, "eval_overall_precision": 0.9312039312039312, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2925, "eval_samples_per_second": 581.258, "eval_steps_per_second": 10.257, "step": 4896 }, { "epoch": 52.0, "grad_norm": 1.0955989360809326, "learning_rate": 2.4e-05, "loss": 0.0127, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9319371727748691, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9081632653061225, "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": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05993056669831276, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9416149068322982, "eval_overall_precision": 0.9381188118811881, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2935, "eval_samples_per_second": 579.129, "eval_steps_per_second": 10.22, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.03147509694099426, "learning_rate": 2.35e-05, "loss": 0.012, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9212121212121213, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.926829268292683, "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": 0.057255420833826065, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9417596034696407, "eval_overall_precision": 0.9359605911330049, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2958, "eval_samples_per_second": 574.635, "eval_steps_per_second": 10.141, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.5938683152198792, "learning_rate": 2.3000000000000003e-05, "loss": 0.0114, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.9417989417989417, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9270833333333334, "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": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.057185981422662735, "eval_overall_accuracy": 0.9893004115226337, "eval_overall_f1": 0.946583850931677, "eval_overall_precision": 0.943069306930693, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2935, "eval_samples_per_second": 579.235, "eval_steps_per_second": 10.222, "step": 5184 }, { "epoch": 55.0, "grad_norm": 0.46015268564224243, "learning_rate": 2.25e-05, "loss": 0.0115, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.9430051813471503, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.91, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9179331306990882, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9263803680981595, "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": 0.057418834418058395, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9440993788819876, "eval_overall_precision": 0.9405940594059405, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.294, "eval_samples_per_second": 578.325, "eval_steps_per_second": 10.206, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.3666340410709381, "learning_rate": 2.2000000000000003e-05, "loss": 0.0106, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9278350515463919, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.927710843373494, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.927710843373494, "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": 0.059970397502183914, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9432098765432098, "eval_overall_precision": 0.9339853300733496, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2937, "eval_samples_per_second": 578.886, "eval_steps_per_second": 10.216, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.050972338765859604, "learning_rate": 2.15e-05, "loss": 0.0111, "step": 5472 }, { "epoch": 57.0, "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": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.05957409366965294, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9382716049382714, "eval_overall_precision": 0.9290953545232273, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2947, "eval_samples_per_second": 576.934, "eval_steps_per_second": 10.181, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.8447536826133728, "learning_rate": 2.1e-05, "loss": 0.0117, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9381443298969072, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.900990099009901, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.9272727272727274, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9329268292682927, "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": 0.06103796139359474, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9455445544554455, "eval_overall_precision": 0.9385749385749386, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2928, "eval_samples_per_second": 580.542, "eval_steps_per_second": 10.245, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.026474952697753906, "learning_rate": 2.05e-05, "loss": 0.0091, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9374999999999999, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9216867469879518, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9216867469879518, "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": 0.05952005833387375, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9430693069306931, "eval_overall_precision": 0.9361179361179361, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2954, "eval_samples_per_second": 575.444, "eval_steps_per_second": 10.155, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.16579730808734894, "learning_rate": 2e-05, "loss": 0.0095, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.9374999999999999, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.967741935483871, "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": 0.0566517598927021, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9456790123456791, "eval_overall_precision": 0.9364303178484108, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2939, "eval_samples_per_second": 578.464, "eval_steps_per_second": 10.208, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.3788284659385681, "learning_rate": 1.9500000000000003e-05, "loss": 0.011, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.946236559139785, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.946236559139785, "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": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.058977656066417694, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9455445544554455, "eval_overall_precision": 0.9385749385749386, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2936, "eval_samples_per_second": 579.115, "eval_steps_per_second": 10.22, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.4230453073978424, "learning_rate": 1.9e-05, "loss": 0.0109, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.911854103343465, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9202453987730062, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06497975438833237, "eval_overall_accuracy": 0.9865569272976681, "eval_overall_f1": 0.9354838709677419, "eval_overall_precision": 0.9308641975308642, "eval_overall_recall": 0.940149625935162, "eval_runtime": 0.2935, "eval_samples_per_second": 579.189, "eval_steps_per_second": 10.221, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.7139930725097656, "learning_rate": 1.85e-05, "loss": 0.0101, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.9430051813471503, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.91, "eval_LOCATION_recall": 0.978494623655914, "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": 0.06298674643039703, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9456790123456791, "eval_overall_precision": 0.9364303178484108, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2927, "eval_samples_per_second": 580.719, "eval_steps_per_second": 10.248, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.04391416534781456, "learning_rate": 1.8e-05, "loss": 0.0098, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.9312169312169313, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.9253731343283581, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9171597633136095, "eval_ORGANIZATION_recall": 0.9337349397590361, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06049938127398491, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9442379182156134, "eval_overall_precision": 0.9384236453201971, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2951, "eval_samples_per_second": 576.162, "eval_steps_per_second": 10.168, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.25877487659454346, "learning_rate": 1.75e-05, "loss": 0.0093, "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": 0.9129129129129129, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.965034965034965, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9583333333333334, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06511997431516647, "eval_overall_accuracy": 0.9865569272976681, "eval_overall_f1": 0.9321824907521579, "eval_overall_precision": 0.9219512195121952, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2926, "eval_samples_per_second": 581.067, "eval_steps_per_second": 10.254, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.6640575528144836, "learning_rate": 1.7000000000000003e-05, "loss": 0.0089, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9278350515463919, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8910891089108911, "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": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06511464715003967, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9382716049382714, "eval_overall_precision": 0.9290953545232273, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2944, "eval_samples_per_second": 577.425, "eval_steps_per_second": 10.19, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.013872658833861351, "learning_rate": 1.65e-05, "loss": 0.0081, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8737864077669902, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9146341463414634, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9259259259259259, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.06600550562143326, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9381188118811881, "eval_overall_precision": 0.9312039312039312, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2939, "eval_samples_per_second": 578.489, "eval_steps_per_second": 10.209, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.08780185878276825, "learning_rate": 1.6000000000000003e-05, "loss": 0.0093, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9374999999999999, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9212121212121213, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.926829268292683, "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": 0.06460719555616379, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9429280397022332, "eval_overall_precision": 0.9382716049382716, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2939, "eval_samples_per_second": 578.427, "eval_steps_per_second": 10.208, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.06301407516002655, "learning_rate": 1.55e-05, "loss": 0.0079, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.946808510638298, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9368421052631579, "eval_LOCATION_recall": 0.956989247311828, "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": 0.061037153005599976, "eval_overall_accuracy": 0.9895747599451303, "eval_overall_f1": 0.946583850931677, "eval_overall_precision": 0.943069306930693, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2981, "eval_samples_per_second": 570.262, "eval_steps_per_second": 10.063, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.1892562061548233, "learning_rate": 1.5e-05, "loss": 0.0079, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9156626506024096, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9156626506024096, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06376679241657257, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9429280397022332, "eval_overall_precision": 0.9382716049382716, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2938, "eval_samples_per_second": 578.591, "eval_steps_per_second": 10.21, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.08717113733291626, "learning_rate": 1.45e-05, "loss": 0.0077, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9312169312169313, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9166666666666666, "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": 0.9754385964912281, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.972027972027972, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.06333258002996445, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9455445544554455, "eval_overall_precision": 0.9385749385749386, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2931, "eval_samples_per_second": 580.025, "eval_steps_per_second": 10.236, "step": 6816 }, { "epoch": 72.0, "grad_norm": 1.032971739768982, "learning_rate": 1.4000000000000001e-05, "loss": 0.0082, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06669674813747406, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2924, "eval_samples_per_second": 581.48, "eval_steps_per_second": 10.261, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.3119294345378876, "learning_rate": 1.3500000000000001e-05, "loss": 0.0086, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.9216867469879518, "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": 0.06375954300165176, "eval_overall_accuracy": 0.9895747599451303, "eval_overall_f1": 0.9417596034696407, "eval_overall_precision": 0.9359605911330049, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2956, "eval_samples_per_second": 575.026, "eval_steps_per_second": 10.148, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.010013816878199577, "learning_rate": 1.3000000000000001e-05, "loss": 0.0084, "step": 7104 }, { "epoch": 74.0, "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": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.0691668763756752, "eval_overall_accuracy": 0.9873799725651577, "eval_overall_f1": 0.9392812887236679, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2973, "eval_samples_per_second": 571.866, "eval_steps_per_second": 10.092, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.3622521758079529, "learning_rate": 1.25e-05, "loss": 0.0078, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.922279792746114, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.89, "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": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06770838052034378, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2954, "eval_samples_per_second": 575.432, "eval_steps_per_second": 10.155, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.34580254554748535, "learning_rate": 1.2e-05, "loss": 0.0082, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9368421052631579, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9175257731958762, "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": 0.0630577802658081, "eval_overall_accuracy": 0.9898491083676269, "eval_overall_f1": 0.9490683229813663, "eval_overall_precision": 0.9455445544554455, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2934, "eval_samples_per_second": 579.425, "eval_steps_per_second": 10.225, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.07749779522418976, "learning_rate": 1.1500000000000002e-05, "loss": 0.007, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.9214659685863874, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.946236559139785, "eval_ORGANIZATION_f1": 0.91131498470948, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9254658385093167, "eval_ORGANIZATION_recall": 0.8975903614457831, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.07113731652498245, "eval_overall_accuracy": 0.9868312757201646, "eval_overall_f1": 0.9363295880149813, "eval_overall_precision": 0.9375, "eval_overall_recall": 0.9351620947630923, "eval_runtime": 0.2933, "eval_samples_per_second": 579.667, "eval_steps_per_second": 10.229, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.21204520761966705, "learning_rate": 1.1000000000000001e-05, "loss": 0.008, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9393939393939394, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9451219512195121, "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": 0.06449893862009048, "eval_overall_accuracy": 0.9906721536351166, "eval_overall_f1": 0.9540372670807453, "eval_overall_precision": 0.9504950495049505, "eval_overall_recall": 0.9576059850374065, "eval_runtime": 0.2938, "eval_samples_per_second": 578.658, "eval_steps_per_second": 10.212, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.07025783509016037, "learning_rate": 1.05e-05, "loss": 0.0066, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9319371727748691, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9081632653061225, "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": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.062268126755952835, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9490683229813663, "eval_overall_precision": 0.9455445544554455, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2947, "eval_samples_per_second": 576.85, "eval_steps_per_second": 10.18, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.12798094749450684, "learning_rate": 1e-05, "loss": 0.0069, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.920245398773006, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9375, "eval_ORGANIZATION_recall": 0.9036144578313253, "eval_PERSON_f1": 0.9823321554770318, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9858156028368794, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.0696575865149498, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.941469489414695, "eval_overall_precision": 0.9402985074626866, "eval_overall_recall": 0.942643391521197, "eval_runtime": 0.2967, "eval_samples_per_second": 572.922, "eval_steps_per_second": 10.11, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.112739697098732, "learning_rate": 9.5e-06, "loss": 0.0058, "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": 0.9244712990936556, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9272727272727272, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.968421052631579, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.965034965034965, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06682921946048737, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9381188118811881, "eval_overall_precision": 0.9312039312039312, "eval_overall_recall": 0.9451371571072319, "eval_runtime": 0.2946, "eval_samples_per_second": 577.151, "eval_steps_per_second": 10.185, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.13783079385757446, "learning_rate": 9e-06, "loss": 0.0069, "step": 7872 }, { "epoch": 82.0, "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": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.06830577552318573, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9429280397022332, "eval_overall_precision": 0.9382716049382716, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2947, "eval_samples_per_second": 576.946, "eval_steps_per_second": 10.181, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.3780260980129242, "learning_rate": 8.500000000000002e-06, "loss": 0.0068, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.956989247311828, "eval_ORGANIZATION_f1": 0.918429003021148, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9212121212121213, "eval_ORGANIZATION_recall": 0.9156626506024096, "eval_PERSON_f1": 0.9788732394366197, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9788732394366197, "eval_PERSON_recall": 0.9788732394366197, "eval_loss": 0.07063104957342148, "eval_overall_accuracy": 0.9876543209876543, "eval_overall_f1": 0.9417596034696407, "eval_overall_precision": 0.9359605911330049, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.294, "eval_samples_per_second": 578.264, "eval_steps_per_second": 10.205, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.25573885440826416, "learning_rate": 8.000000000000001e-06, "loss": 0.0065, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9305135951661632, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9333333333333333, "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": 0.06899310648441315, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.949194547707559, "eval_overall_precision": 0.9433497536945813, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.294, "eval_samples_per_second": 578.246, "eval_steps_per_second": 10.204, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.07031101733446121, "learning_rate": 7.5e-06, "loss": 0.0066, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.927710843373494, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.927710843373494, "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": 0.0673184022307396, "eval_overall_accuracy": 0.9879286694101509, "eval_overall_f1": 0.9468479604449938, "eval_overall_precision": 0.9387254901960784, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2953, "eval_samples_per_second": 575.681, "eval_steps_per_second": 10.159, "step": 8160 }, { "epoch": 86.0, "grad_norm": 1.1244421005249023, "learning_rate": 7.000000000000001e-06, "loss": 0.0063, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9161676646706587, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9107142857142857, "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": 0.06586689502000809, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9430693069306931, "eval_overall_precision": 0.9361179361179361, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2949, "eval_samples_per_second": 576.492, "eval_steps_per_second": 10.173, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.23708193004131317, "learning_rate": 6.5000000000000004e-06, "loss": 0.0064, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9244712990936556, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9272727272727272, "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": 0.06606775522232056, "eval_overall_accuracy": 0.9895747599451303, "eval_overall_f1": 0.946583850931677, "eval_overall_precision": 0.943069306930693, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2934, "eval_samples_per_second": 579.336, "eval_steps_per_second": 10.224, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.2592020332813263, "learning_rate": 6e-06, "loss": 0.0056, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.924924924924925, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9221556886227545, "eval_ORGANIZATION_recall": 0.927710843373494, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06428737938404083, "eval_overall_accuracy": 0.9898491083676269, "eval_overall_f1": 0.946716232961586, "eval_overall_precision": 0.9408866995073891, "eval_overall_recall": 0.9526184538653366, "eval_runtime": 0.2932, "eval_samples_per_second": 579.761, "eval_steps_per_second": 10.231, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.17049174010753632, "learning_rate": 5.500000000000001e-06, "loss": 0.0065, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9374999999999999, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06466696411371231, "eval_overall_accuracy": 0.9893004115226337, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2968, "eval_samples_per_second": 572.687, "eval_steps_per_second": 10.106, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.5506210327148438, "learning_rate": 5e-06, "loss": 0.0062, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9244712990936556, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9272727272727272, "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": 0.06439953297376633, "eval_overall_accuracy": 0.9895747599451303, "eval_overall_f1": 0.946583850931677, "eval_overall_precision": 0.943069306930693, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2948, "eval_samples_per_second": 576.759, "eval_steps_per_second": 10.178, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.12727555632591248, "learning_rate": 4.5e-06, "loss": 0.0058, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9216867469879518, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9216867469879518, "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": 0.0659627914428711, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2937, "eval_samples_per_second": 578.886, "eval_steps_per_second": 10.216, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.08370932936668396, "learning_rate": 4.000000000000001e-06, "loss": 0.0051, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9479166666666667, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9191919191919192, "eval_LOCATION_recall": 0.978494623655914, "eval_ORGANIZATION_f1": 0.927710843373494, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.927710843373494, "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": 0.06474878638982773, "eval_overall_accuracy": 0.9887517146776406, "eval_overall_f1": 0.9480198019801981, "eval_overall_precision": 0.941031941031941, "eval_overall_recall": 0.9551122194513716, "eval_runtime": 0.2932, "eval_samples_per_second": 579.905, "eval_steps_per_second": 10.234, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.023883234709501266, "learning_rate": 3.5000000000000004e-06, "loss": 0.0048, "step": 8928 }, { "epoch": 93.0, "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": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06717947870492935, "eval_overall_accuracy": 0.9882030178326474, "eval_overall_f1": 0.9371146732429101, "eval_overall_precision": 0.926829268292683, "eval_overall_recall": 0.9476309226932669, "eval_runtime": 0.2947, "eval_samples_per_second": 576.916, "eval_steps_per_second": 10.181, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.8206273317337036, "learning_rate": 3e-06, "loss": 0.0058, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.9374999999999999, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06682807952165604, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9430693069306931, "eval_overall_precision": 0.9361179361179361, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2937, "eval_samples_per_second": 578.863, "eval_steps_per_second": 10.215, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.07877936959266663, "learning_rate": 2.5e-06, "loss": 0.0067, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.9374999999999999, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.971830985915493, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.971830985915493, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06732748448848724, "eval_overall_accuracy": 0.988477366255144, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2935, "eval_samples_per_second": 579.128, "eval_steps_per_second": 10.22, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.03156942501664162, "learning_rate": 2.0000000000000003e-06, "loss": 0.0051, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.9473684210526316, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9278350515463918, "eval_LOCATION_recall": 0.967741935483871, "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": 0.06718067824840546, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.946583850931677, "eval_overall_precision": 0.943069306930693, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2918, "eval_samples_per_second": 582.535, "eval_steps_per_second": 10.28, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.007113738916814327, "learning_rate": 1.5e-06, "loss": 0.005, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06768758594989777, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2919, "eval_samples_per_second": 582.295, "eval_steps_per_second": 10.276, "step": 9312 }, { "epoch": 98.0, "grad_norm": 0.05270858108997345, "learning_rate": 1.0000000000000002e-06, "loss": 0.0062, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9189189189189191, "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": 0.0679173544049263, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2941, "eval_samples_per_second": 578.041, "eval_steps_per_second": 10.201, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.07660820335149765, "learning_rate": 5.000000000000001e-07, "loss": 0.0053, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.06781742721796036, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2949, "eval_samples_per_second": 576.383, "eval_steps_per_second": 10.171, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.8095541596412659, "learning_rate": 0.0, "loss": 0.0046, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 93, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.967741935483871, "eval_ORGANIZATION_f1": 0.9189189189189191, "eval_ORGANIZATION_number": 166, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9216867469879518, "eval_PERSON_f1": 0.9752650176678446, "eval_PERSON_number": 142, "eval_PERSON_precision": 0.9787234042553191, "eval_PERSON_recall": 0.971830985915493, "eval_loss": 0.067685067653656, "eval_overall_accuracy": 0.9890260631001372, "eval_overall_f1": 0.9419035846724352, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.9501246882793017, "eval_runtime": 0.2957, "eval_samples_per_second": 574.904, "eval_steps_per_second": 10.145, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 4208466347875278.0, "train_loss": 0.03501902000978589, "train_runtime": 539.9434, "train_samples_per_second": 283.548, "train_steps_per_second": 17.78 } ], "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 }