{
"best_metric": 0.9184872813625782,
"best_model_checkpoint": "../out/onto_final/ner-roberta-qa4-querytype4-span-char-pattern3-subtext-dim50/checkpoints/checkpoint-288000",
"epoch": 230.58446757405926,
"global_step": 288000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.64,
"learning_rate": 9.978649586335736e-06,
"loss": 0.0794,
"step": 800
},
{
"epoch": 0.64,
"eval_loss": 0.03532102704048157,
"eval_micro_f1": 0.8791638865358467,
"eval_runtime": 57.079,
"eval_samples_per_second": 149.407,
"step": 800
},
{
"epoch": 1.28,
"learning_rate": 9.957299172671472e-06,
"loss": 0.0358,
"step": 1600
},
{
"epoch": 1.28,
"eval_loss": 0.03331479802727699,
"eval_micro_f1": 0.8892428863535353,
"eval_runtime": 61.3385,
"eval_samples_per_second": 139.032,
"step": 1600
},
{
"epoch": 1.92,
"learning_rate": 9.935948759007207e-06,
"loss": 0.0309,
"step": 2400
},
{
"epoch": 1.92,
"eval_loss": 0.030986160039901733,
"eval_micro_f1": 0.8977142347839291,
"eval_runtime": 56.7577,
"eval_samples_per_second": 150.253,
"step": 2400
},
{
"epoch": 2.56,
"learning_rate": 9.914598345342942e-06,
"loss": 0.0265,
"step": 3200
},
{
"epoch": 2.56,
"eval_loss": 0.03181663900613785,
"eval_micro_f1": 0.9019206824095267,
"eval_runtime": 57.1062,
"eval_samples_per_second": 149.336,
"step": 3200
},
{
"epoch": 3.2,
"learning_rate": 9.893247931678677e-06,
"loss": 0.0238,
"step": 4000
},
{
"epoch": 3.2,
"eval_loss": 0.0325116403400898,
"eval_micro_f1": 0.9057191136154928,
"eval_runtime": 57.5307,
"eval_samples_per_second": 148.234,
"step": 4000
},
{
"epoch": 3.84,
"learning_rate": 9.871897518014413e-06,
"loss": 0.0217,
"step": 4800
},
{
"epoch": 3.84,
"eval_loss": 0.03316809609532356,
"eval_micro_f1": 0.9071629899701918,
"eval_runtime": 57.2457,
"eval_samples_per_second": 148.972,
"step": 4800
},
{
"epoch": 4.48,
"learning_rate": 9.850547104350148e-06,
"loss": 0.0179,
"step": 5600
},
{
"epoch": 4.48,
"eval_loss": 0.03316080570220947,
"eval_micro_f1": 0.9078562750766709,
"eval_runtime": 59.8531,
"eval_samples_per_second": 142.482,
"step": 5600
},
{
"epoch": 5.12,
"learning_rate": 9.829196690685883e-06,
"loss": 0.0178,
"step": 6400
},
{
"epoch": 5.12,
"eval_loss": 0.03498595580458641,
"eval_micro_f1": 0.9059721593562468,
"eval_runtime": 57.208,
"eval_samples_per_second": 149.07,
"step": 6400
},
{
"epoch": 5.76,
"learning_rate": 9.807846277021618e-06,
"loss": 0.0151,
"step": 7200
},
{
"epoch": 5.76,
"eval_loss": 0.03542535379528999,
"eval_micro_f1": 0.9050514076358719,
"eval_runtime": 57.0443,
"eval_samples_per_second": 149.498,
"step": 7200
},
{
"epoch": 6.41,
"learning_rate": 9.786495863357354e-06,
"loss": 0.0137,
"step": 8000
},
{
"epoch": 6.41,
"eval_loss": 0.039753422141075134,
"eval_micro_f1": 0.9075421467937833,
"eval_runtime": 57.0434,
"eval_samples_per_second": 149.5,
"step": 8000
},
{
"epoch": 7.05,
"learning_rate": 9.765145449693089e-06,
"loss": 0.0128,
"step": 8800
},
{
"epoch": 7.05,
"eval_loss": 0.04039251431822777,
"eval_micro_f1": 0.9071243402526367,
"eval_runtime": 60.8774,
"eval_samples_per_second": 140.085,
"step": 8800
},
{
"epoch": 7.69,
"learning_rate": 9.743795036028824e-06,
"loss": 0.0108,
"step": 9600
},
{
"epoch": 7.69,
"eval_loss": 0.040012676268815994,
"eval_micro_f1": 0.9102099922587893,
"eval_runtime": 57.1723,
"eval_samples_per_second": 149.163,
"step": 9600
},
{
"epoch": 8.33,
"learning_rate": 9.72244462236456e-06,
"loss": 0.0107,
"step": 10400
},
{
"epoch": 8.33,
"eval_loss": 0.04479105398058891,
"eval_micro_f1": 0.9105807544365998,
"eval_runtime": 57.4491,
"eval_samples_per_second": 148.444,
"step": 10400
},
{
"epoch": 8.97,
"learning_rate": 9.701094208700295e-06,
"loss": 0.0098,
"step": 11200
},
{
"epoch": 8.97,
"eval_loss": 0.04202866554260254,
"eval_micro_f1": 0.9104168537359922,
"eval_runtime": 59.4767,
"eval_samples_per_second": 143.384,
"step": 11200
},
{
"epoch": 9.61,
"learning_rate": 9.67974379503603e-06,
"loss": 0.0082,
"step": 12000
},
{
"epoch": 9.61,
"eval_loss": 0.04732278734445572,
"eval_micro_f1": 0.9088144532363872,
"eval_runtime": 59.4788,
"eval_samples_per_second": 143.379,
"step": 12000
},
{
"epoch": 10.25,
"learning_rate": 9.658393381371765e-06,
"loss": 0.0078,
"step": 12800
},
{
"epoch": 10.25,
"eval_loss": 0.05117360129952431,
"eval_micro_f1": 0.9097411591405486,
"eval_runtime": 59.2888,
"eval_samples_per_second": 143.838,
"step": 12800
},
{
"epoch": 10.89,
"learning_rate": 9.637042967707501e-06,
"loss": 0.0072,
"step": 13600
},
{
"epoch": 10.89,
"eval_loss": 0.049624327570199966,
"eval_micro_f1": 0.9090827534806584,
"eval_runtime": 59.1661,
"eval_samples_per_second": 144.137,
"step": 13600
},
{
"epoch": 11.53,
"learning_rate": 9.615692554043236e-06,
"loss": 0.0062,
"step": 14400
},
{
"epoch": 11.53,
"eval_loss": 0.05105595663189888,
"eval_micro_f1": 0.9028098874602113,
"eval_runtime": 62.0832,
"eval_samples_per_second": 137.364,
"step": 14400
},
{
"epoch": 12.17,
"learning_rate": 9.59434214037897e-06,
"loss": 0.0063,
"step": 15200
},
{
"epoch": 12.17,
"eval_loss": 0.05476413667201996,
"eval_micro_f1": 0.9109142415289799,
"eval_runtime": 56.7241,
"eval_samples_per_second": 150.342,
"step": 15200
},
{
"epoch": 12.81,
"learning_rate": 9.572991726714706e-06,
"loss": 0.0055,
"step": 16000
},
{
"epoch": 12.81,
"eval_loss": 0.054592572152614594,
"eval_micro_f1": 0.9079886833204951,
"eval_runtime": 57.0367,
"eval_samples_per_second": 149.518,
"step": 16000
},
{
"epoch": 13.45,
"learning_rate": 9.551641313050442e-06,
"loss": 0.005,
"step": 16800
},
{
"epoch": 13.45,
"eval_loss": 0.060022544115781784,
"eval_micro_f1": 0.9115664716580582,
"eval_runtime": 56.9948,
"eval_samples_per_second": 149.628,
"step": 16800
},
{
"epoch": 14.09,
"learning_rate": 9.530290899386177e-06,
"loss": 0.0047,
"step": 17600
},
{
"epoch": 14.09,
"eval_loss": 0.05833619460463524,
"eval_micro_f1": 0.9108928566428472,
"eval_runtime": 57.2197,
"eval_samples_per_second": 149.04,
"step": 17600
},
{
"epoch": 14.73,
"learning_rate": 9.508940485721912e-06,
"loss": 0.0042,
"step": 18400
},
{
"epoch": 14.73,
"eval_loss": 0.05999485030770302,
"eval_micro_f1": 0.9137705651974638,
"eval_runtime": 57.0857,
"eval_samples_per_second": 149.389,
"step": 18400
},
{
"epoch": 15.37,
"learning_rate": 9.487590072057647e-06,
"loss": 0.0042,
"step": 19200
},
{
"epoch": 15.37,
"eval_loss": 0.059445686638355255,
"eval_micro_f1": 0.9135040896166442,
"eval_runtime": 59.2754,
"eval_samples_per_second": 143.871,
"step": 19200
},
{
"epoch": 16.01,
"learning_rate": 9.466239658393383e-06,
"loss": 0.004,
"step": 20000
},
{
"epoch": 16.01,
"eval_loss": 0.06065479665994644,
"eval_micro_f1": 0.9133491553819155,
"eval_runtime": 61.7392,
"eval_samples_per_second": 138.129,
"step": 20000
},
{
"epoch": 16.65,
"learning_rate": 9.444889244729118e-06,
"loss": 0.0033,
"step": 20800
},
{
"epoch": 16.65,
"eval_loss": 0.0621187798678875,
"eval_micro_f1": 0.9082957735193624,
"eval_runtime": 57.2265,
"eval_samples_per_second": 149.022,
"step": 20800
},
{
"epoch": 17.29,
"learning_rate": 9.423538831064853e-06,
"loss": 0.0034,
"step": 21600
},
{
"epoch": 17.29,
"eval_loss": 0.06206576153635979,
"eval_micro_f1": 0.9121545715062239,
"eval_runtime": 59.5342,
"eval_samples_per_second": 143.245,
"step": 21600
},
{
"epoch": 17.93,
"learning_rate": 9.40218841740059e-06,
"loss": 0.0034,
"step": 22400
},
{
"epoch": 17.93,
"eval_loss": 0.06079624593257904,
"eval_micro_f1": 0.9086459301723967,
"eval_runtime": 57.0158,
"eval_samples_per_second": 149.573,
"step": 22400
},
{
"epoch": 18.57,
"learning_rate": 9.380838003736324e-06,
"loss": 0.0029,
"step": 23200
},
{
"epoch": 18.57,
"eval_loss": 0.06456855684518814,
"eval_micro_f1": 0.9081564240809833,
"eval_runtime": 57.1215,
"eval_samples_per_second": 149.296,
"step": 23200
},
{
"epoch": 19.22,
"learning_rate": 9.359487590072059e-06,
"loss": 0.0027,
"step": 24000
},
{
"epoch": 19.22,
"eval_loss": 0.06243640184402466,
"eval_micro_f1": 0.9079512888982412,
"eval_runtime": 59.6186,
"eval_samples_per_second": 143.043,
"step": 24000
},
{
"epoch": 19.86,
"learning_rate": 9.338137176407794e-06,
"loss": 0.0027,
"step": 24800
},
{
"epoch": 19.86,
"eval_loss": 0.06747569888830185,
"eval_micro_f1": 0.9114344735859866,
"eval_runtime": 57.0114,
"eval_samples_per_second": 149.584,
"step": 24800
},
{
"epoch": 20.5,
"learning_rate": 9.31678676274353e-06,
"loss": 0.0024,
"step": 25600
},
{
"epoch": 20.5,
"eval_loss": 0.06775301694869995,
"eval_micro_f1": 0.9094236859425605,
"eval_runtime": 64.1251,
"eval_samples_per_second": 132.99,
"step": 25600
},
{
"epoch": 21.14,
"learning_rate": 9.295436349079265e-06,
"loss": 0.0023,
"step": 26400
},
{
"epoch": 21.14,
"eval_loss": 0.06704603135585785,
"eval_micro_f1": 0.9129458264696592,
"eval_runtime": 57.1031,
"eval_samples_per_second": 149.344,
"step": 26400
},
{
"epoch": 21.78,
"learning_rate": 9.274085935415e-06,
"loss": 0.0022,
"step": 27200
},
{
"epoch": 21.78,
"eval_loss": 0.06668079644441605,
"eval_micro_f1": 0.9119217476042184,
"eval_runtime": 59.3802,
"eval_samples_per_second": 143.617,
"step": 27200
},
{
"epoch": 22.42,
"learning_rate": 9.252735521750735e-06,
"loss": 0.0021,
"step": 28000
},
{
"epoch": 22.42,
"eval_loss": 0.07310529053211212,
"eval_micro_f1": 0.9129456242748618,
"eval_runtime": 59.3722,
"eval_samples_per_second": 143.636,
"step": 28000
},
{
"epoch": 23.06,
"learning_rate": 9.231385108086471e-06,
"loss": 0.002,
"step": 28800
},
{
"epoch": 23.06,
"eval_loss": 0.06730842590332031,
"eval_micro_f1": 0.9117686794080435,
"eval_runtime": 57.1302,
"eval_samples_per_second": 149.273,
"step": 28800
},
{
"epoch": 23.7,
"learning_rate": 9.210034694422206e-06,
"loss": 0.0019,
"step": 29600
},
{
"epoch": 23.7,
"eval_loss": 0.06992793083190918,
"eval_micro_f1": 0.9109914754981487,
"eval_runtime": 56.9315,
"eval_samples_per_second": 149.794,
"step": 29600
},
{
"epoch": 24.34,
"learning_rate": 9.18868428075794e-06,
"loss": 0.0017,
"step": 30400
},
{
"epoch": 24.34,
"eval_loss": 0.07501634210348129,
"eval_micro_f1": 0.9093111953716045,
"eval_runtime": 57.3844,
"eval_samples_per_second": 148.612,
"step": 30400
},
{
"epoch": 24.98,
"learning_rate": 9.167333867093676e-06,
"loss": 0.0018,
"step": 31200
},
{
"epoch": 24.98,
"eval_loss": 0.07312783598899841,
"eval_micro_f1": 0.9114410944100688,
"eval_runtime": 64.5682,
"eval_samples_per_second": 132.077,
"step": 31200
},
{
"epoch": 25.62,
"learning_rate": 9.145983453429412e-06,
"loss": 0.0015,
"step": 32000
},
{
"epoch": 25.62,
"eval_loss": 0.07030045986175537,
"eval_micro_f1": 0.9128145109101065,
"eval_runtime": 59.4137,
"eval_samples_per_second": 143.536,
"step": 32000
},
{
"epoch": 26.26,
"learning_rate": 9.124633039765147e-06,
"loss": 0.0016,
"step": 32800
},
{
"epoch": 26.26,
"eval_loss": 0.07460073381662369,
"eval_micro_f1": 0.9134680281189244,
"eval_runtime": 57.0981,
"eval_samples_per_second": 149.357,
"step": 32800
},
{
"epoch": 26.9,
"learning_rate": 9.103282626100882e-06,
"loss": 0.0016,
"step": 33600
},
{
"epoch": 26.9,
"eval_loss": 0.07456312328577042,
"eval_micro_f1": 0.9089120417125551,
"eval_runtime": 57.0888,
"eval_samples_per_second": 149.381,
"step": 33600
},
{
"epoch": 27.54,
"learning_rate": 9.081932212436616e-06,
"loss": 0.0015,
"step": 34400
},
{
"epoch": 27.54,
"eval_loss": 0.07069224119186401,
"eval_micro_f1": 0.9102228633454481,
"eval_runtime": 56.8397,
"eval_samples_per_second": 150.036,
"step": 34400
},
{
"epoch": 28.18,
"learning_rate": 9.060581798772351e-06,
"loss": 0.0015,
"step": 35200
},
{
"epoch": 28.18,
"eval_loss": 0.07040959596633911,
"eval_micro_f1": 0.9124887685924771,
"eval_runtime": 57.015,
"eval_samples_per_second": 149.575,
"step": 35200
},
{
"epoch": 28.82,
"learning_rate": 9.039231385108088e-06,
"loss": 0.0015,
"step": 36000
},
{
"epoch": 28.82,
"eval_loss": 0.07187571376562119,
"eval_micro_f1": 0.9107327368196703,
"eval_runtime": 59.5606,
"eval_samples_per_second": 143.182,
"step": 36000
},
{
"epoch": 29.46,
"learning_rate": 9.017880971443823e-06,
"loss": 0.0012,
"step": 36800
},
{
"epoch": 29.46,
"eval_loss": 0.07232771068811417,
"eval_micro_f1": 0.9132399452748835,
"eval_runtime": 61.6472,
"eval_samples_per_second": 138.336,
"step": 36800
},
{
"epoch": 30.1,
"learning_rate": 8.996530557779557e-06,
"loss": 0.0013,
"step": 37600
},
{
"epoch": 30.1,
"eval_loss": 0.07187189161777496,
"eval_micro_f1": 0.9100192586928618,
"eval_runtime": 57.0806,
"eval_samples_per_second": 149.403,
"step": 37600
},
{
"epoch": 30.74,
"learning_rate": 8.975180144115292e-06,
"loss": 0.0012,
"step": 38400
},
{
"epoch": 30.74,
"eval_loss": 0.0734463706612587,
"eval_micro_f1": 0.9114715998038618,
"eval_runtime": 56.9486,
"eval_samples_per_second": 149.749,
"step": 38400
},
{
"epoch": 31.39,
"learning_rate": 8.953829730451029e-06,
"loss": 0.0012,
"step": 39200
},
{
"epoch": 31.39,
"eval_loss": 0.07252092659473419,
"eval_micro_f1": 0.9121512146214378,
"eval_runtime": 59.3108,
"eval_samples_per_second": 143.785,
"step": 39200
},
{
"epoch": 32.03,
"learning_rate": 8.932479316786764e-06,
"loss": 0.0012,
"step": 40000
},
{
"epoch": 32.03,
"eval_loss": 0.06957610696554184,
"eval_micro_f1": 0.9138822677915149,
"eval_runtime": 56.7133,
"eval_samples_per_second": 150.37,
"step": 40000
},
{
"epoch": 32.67,
"learning_rate": 8.911128903122498e-06,
"loss": 0.0012,
"step": 40800
},
{
"epoch": 32.67,
"eval_loss": 0.07325581461191177,
"eval_micro_f1": 0.9099232524957666,
"eval_runtime": 59.2668,
"eval_samples_per_second": 143.892,
"step": 40800
},
{
"epoch": 33.31,
"learning_rate": 8.889778489458233e-06,
"loss": 0.0012,
"step": 41600
},
{
"epoch": 33.31,
"eval_loss": 0.07478074729442596,
"eval_micro_f1": 0.911550561297685,
"eval_runtime": 57.2741,
"eval_samples_per_second": 148.898,
"step": 41600
},
{
"epoch": 33.95,
"learning_rate": 8.86842807579397e-06,
"loss": 0.0011,
"step": 42400
},
{
"epoch": 33.95,
"eval_loss": 0.0724610686302185,
"eval_micro_f1": 0.911983507611451,
"eval_runtime": 59.4574,
"eval_samples_per_second": 143.43,
"step": 42400
},
{
"epoch": 34.59,
"learning_rate": 8.847077662129704e-06,
"loss": 0.001,
"step": 43200
},
{
"epoch": 34.59,
"eval_loss": 0.0751844123005867,
"eval_micro_f1": 0.909139880986741,
"eval_runtime": 59.4779,
"eval_samples_per_second": 143.381,
"step": 43200
},
{
"epoch": 35.23,
"learning_rate": 8.82572724846544e-06,
"loss": 0.0011,
"step": 44000
},
{
"epoch": 35.23,
"eval_loss": 0.0745643824338913,
"eval_micro_f1": 0.9135747208892122,
"eval_runtime": 56.9805,
"eval_samples_per_second": 149.665,
"step": 44000
},
{
"epoch": 35.87,
"learning_rate": 8.804376834801174e-06,
"loss": 0.001,
"step": 44800
},
{
"epoch": 35.87,
"eval_loss": 0.07720115780830383,
"eval_micro_f1": 0.9133405244909418,
"eval_runtime": 59.1924,
"eval_samples_per_second": 144.072,
"step": 44800
},
{
"epoch": 36.51,
"learning_rate": 8.78302642113691e-06,
"loss": 0.0011,
"step": 45600
},
{
"epoch": 36.51,
"eval_loss": 0.07475907355546951,
"eval_micro_f1": 0.912073137440258,
"eval_runtime": 59.4341,
"eval_samples_per_second": 143.487,
"step": 45600
},
{
"epoch": 37.15,
"learning_rate": 8.761676007472645e-06,
"loss": 0.0011,
"step": 46400
},
{
"epoch": 37.15,
"eval_loss": 0.0729161724448204,
"eval_micro_f1": 0.913258426466224,
"eval_runtime": 57.0175,
"eval_samples_per_second": 149.568,
"step": 46400
},
{
"epoch": 37.79,
"learning_rate": 8.74032559380838e-06,
"loss": 0.0011,
"step": 47200
},
{
"epoch": 37.79,
"eval_loss": 0.07795717567205429,
"eval_micro_f1": 0.9141856387293588,
"eval_runtime": 57.304,
"eval_samples_per_second": 148.82,
"step": 47200
},
{
"epoch": 38.43,
"learning_rate": 8.718975180144115e-06,
"loss": 0.0011,
"step": 48000
},
{
"epoch": 38.43,
"eval_loss": 0.07795872539281845,
"eval_micro_f1": 0.9130454275460468,
"eval_runtime": 64.3755,
"eval_samples_per_second": 132.473,
"step": 48000
},
{
"epoch": 39.07,
"learning_rate": 8.697624766479852e-06,
"loss": 0.0009,
"step": 48800
},
{
"epoch": 39.07,
"eval_loss": 0.07526353746652603,
"eval_micro_f1": 0.9140677580819473,
"eval_runtime": 59.1692,
"eval_samples_per_second": 144.129,
"step": 48800
},
{
"epoch": 39.71,
"learning_rate": 8.676274352815586e-06,
"loss": 0.0009,
"step": 49600
},
{
"epoch": 39.71,
"eval_loss": 0.07573703676462173,
"eval_micro_f1": 0.9124100714423757,
"eval_runtime": 59.248,
"eval_samples_per_second": 143.937,
"step": 49600
},
{
"epoch": 40.35,
"learning_rate": 8.654923939151321e-06,
"loss": 0.0009,
"step": 50400
},
{
"epoch": 40.35,
"eval_loss": 0.07883565127849579,
"eval_micro_f1": 0.914203445758981,
"eval_runtime": 56.8891,
"eval_samples_per_second": 149.906,
"step": 50400
},
{
"epoch": 40.99,
"learning_rate": 8.633573525487056e-06,
"loss": 0.0009,
"step": 51200
},
{
"epoch": 40.99,
"eval_loss": 0.076920285820961,
"eval_micro_f1": 0.9128574640349378,
"eval_runtime": 57.1513,
"eval_samples_per_second": 149.218,
"step": 51200
},
{
"epoch": 41.63,
"learning_rate": 8.612223111822793e-06,
"loss": 0.0009,
"step": 52000
},
{
"epoch": 41.63,
"eval_loss": 0.08013898879289627,
"eval_micro_f1": 0.9134982263707027,
"eval_runtime": 56.7548,
"eval_samples_per_second": 150.26,
"step": 52000
},
{
"epoch": 42.27,
"learning_rate": 8.590872698158527e-06,
"loss": 0.0008,
"step": 52800
},
{
"epoch": 42.27,
"eval_loss": 0.08157988637685776,
"eval_micro_f1": 0.9116446034936594,
"eval_runtime": 57.0043,
"eval_samples_per_second": 149.603,
"step": 52800
},
{
"epoch": 42.91,
"learning_rate": 8.569522284494262e-06,
"loss": 0.0011,
"step": 53600
},
{
"epoch": 42.91,
"eval_loss": 0.07772978395223618,
"eval_micro_f1": 0.9130864636379873,
"eval_runtime": 62.6,
"eval_samples_per_second": 136.23,
"step": 53600
},
{
"epoch": 43.55,
"learning_rate": 8.548171870829997e-06,
"loss": 0.0008,
"step": 54400
},
{
"epoch": 43.55,
"eval_loss": 0.0834282636642456,
"eval_micro_f1": 0.9114567082028342,
"eval_runtime": 57.0663,
"eval_samples_per_second": 149.44,
"step": 54400
},
{
"epoch": 44.2,
"learning_rate": 8.526821457165733e-06,
"loss": 0.0007,
"step": 55200
},
{
"epoch": 44.2,
"eval_loss": 0.08250031620264053,
"eval_micro_f1": 0.9122964420103737,
"eval_runtime": 57.2691,
"eval_samples_per_second": 148.911,
"step": 55200
},
{
"epoch": 44.84,
"learning_rate": 8.505471043501468e-06,
"loss": 0.0009,
"step": 56000
},
{
"epoch": 44.84,
"eval_loss": 0.07949114590883255,
"eval_micro_f1": 0.9116998780479013,
"eval_runtime": 59.7399,
"eval_samples_per_second": 142.752,
"step": 56000
},
{
"epoch": 45.48,
"learning_rate": 8.484120629837203e-06,
"loss": 0.0007,
"step": 56800
},
{
"epoch": 45.48,
"eval_loss": 0.07564416527748108,
"eval_micro_f1": 0.9108661412322101,
"eval_runtime": 56.9113,
"eval_samples_per_second": 149.847,
"step": 56800
},
{
"epoch": 46.12,
"learning_rate": 8.462770216172938e-06,
"loss": 0.0008,
"step": 57600
},
{
"epoch": 46.12,
"eval_loss": 0.0804159939289093,
"eval_micro_f1": 0.9148878676436679,
"eval_runtime": 59.2174,
"eval_samples_per_second": 144.012,
"step": 57600
},
{
"epoch": 46.76,
"learning_rate": 8.441419802508674e-06,
"loss": 0.0007,
"step": 58400
},
{
"epoch": 46.76,
"eval_loss": 0.08167388290166855,
"eval_micro_f1": 0.9112607924116305,
"eval_runtime": 56.9517,
"eval_samples_per_second": 149.741,
"step": 58400
},
{
"epoch": 47.4,
"learning_rate": 8.42006938884441e-06,
"loss": 0.0007,
"step": 59200
},
{
"epoch": 47.4,
"eval_loss": 0.08196252584457397,
"eval_micro_f1": 0.913084489460857,
"eval_runtime": 64.1271,
"eval_samples_per_second": 132.986,
"step": 59200
},
{
"epoch": 48.04,
"learning_rate": 8.398718975180144e-06,
"loss": 0.0006,
"step": 60000
},
{
"epoch": 48.04,
"eval_loss": 0.08966843038797379,
"eval_micro_f1": 0.90746884196386,
"eval_runtime": 57.0865,
"eval_samples_per_second": 149.387,
"step": 60000
},
{
"epoch": 48.68,
"learning_rate": 8.377368561515879e-06,
"loss": 0.0008,
"step": 60800
},
{
"epoch": 48.68,
"eval_loss": 0.08180336654186249,
"eval_micro_f1": 0.9101485586551468,
"eval_runtime": 61.6521,
"eval_samples_per_second": 138.325,
"step": 60800
},
{
"epoch": 49.32,
"learning_rate": 8.356018147851615e-06,
"loss": 0.0008,
"step": 61600
},
{
"epoch": 49.32,
"eval_loss": 0.08304371684789658,
"eval_micro_f1": 0.9103984272291247,
"eval_runtime": 59.2356,
"eval_samples_per_second": 143.967,
"step": 61600
},
{
"epoch": 49.96,
"learning_rate": 8.33466773418735e-06,
"loss": 0.0008,
"step": 62400
},
{
"epoch": 49.96,
"eval_loss": 0.0812276303768158,
"eval_micro_f1": 0.9142780239919369,
"eval_runtime": 57.3079,
"eval_samples_per_second": 148.81,
"step": 62400
},
{
"epoch": 50.6,
"learning_rate": 8.313317320523085e-06,
"loss": 0.0005,
"step": 63200
},
{
"epoch": 50.6,
"eval_loss": 0.08474335819482803,
"eval_micro_f1": 0.9113799278153659,
"eval_runtime": 57.0712,
"eval_samples_per_second": 149.428,
"step": 63200
},
{
"epoch": 51.24,
"learning_rate": 8.291966906858822e-06,
"loss": 0.0007,
"step": 64000
},
{
"epoch": 51.24,
"eval_loss": 0.08282973617315292,
"eval_micro_f1": 0.9125434120658837,
"eval_runtime": 57.2782,
"eval_samples_per_second": 148.887,
"step": 64000
},
{
"epoch": 51.88,
"learning_rate": 8.270616493194556e-06,
"loss": 0.0006,
"step": 64800
},
{
"epoch": 51.88,
"eval_loss": 0.07545238733291626,
"eval_micro_f1": 0.9108554991404801,
"eval_runtime": 60.2593,
"eval_samples_per_second": 141.522,
"step": 64800
},
{
"epoch": 52.52,
"learning_rate": 8.249266079530291e-06,
"loss": 0.0007,
"step": 65600
},
{
"epoch": 52.52,
"eval_loss": 0.07882095873355865,
"eval_micro_f1": 0.9138857524915636,
"eval_runtime": 57.0428,
"eval_samples_per_second": 149.502,
"step": 65600
},
{
"epoch": 53.16,
"learning_rate": 8.227915665866026e-06,
"loss": 0.0006,
"step": 66400
},
{
"epoch": 53.16,
"eval_loss": 0.08688221126794815,
"eval_micro_f1": 0.9132072083226869,
"eval_runtime": 63.3368,
"eval_samples_per_second": 134.645,
"step": 66400
},
{
"epoch": 53.8,
"learning_rate": 8.206565252201762e-06,
"loss": 0.0006,
"step": 67200
},
{
"epoch": 53.8,
"eval_loss": 0.08366230875253677,
"eval_micro_f1": 0.9120588362190817,
"eval_runtime": 57.1166,
"eval_samples_per_second": 149.309,
"step": 67200
},
{
"epoch": 54.44,
"learning_rate": 8.185214838537497e-06,
"loss": 0.0006,
"step": 68000
},
{
"epoch": 54.44,
"eval_loss": 0.08101100474596024,
"eval_micro_f1": 0.9122144265551545,
"eval_runtime": 57.0033,
"eval_samples_per_second": 149.605,
"step": 68000
},
{
"epoch": 55.08,
"learning_rate": 8.163864424873232e-06,
"loss": 0.0006,
"step": 68800
},
{
"epoch": 55.08,
"eval_loss": 0.0850592628121376,
"eval_micro_f1": 0.9114218067988622,
"eval_runtime": 57.4412,
"eval_samples_per_second": 148.465,
"step": 68800
},
{
"epoch": 55.72,
"learning_rate": 8.142514011208967e-06,
"loss": 0.0007,
"step": 69600
},
{
"epoch": 55.72,
"eval_loss": 0.08296786993741989,
"eval_micro_f1": 0.9129498334764293,
"eval_runtime": 56.8369,
"eval_samples_per_second": 150.043,
"step": 69600
},
{
"epoch": 56.37,
"learning_rate": 8.121163597544703e-06,
"loss": 0.0006,
"step": 70400
},
{
"epoch": 56.37,
"eval_loss": 0.08281563967466354,
"eval_micro_f1": 0.9138805362319787,
"eval_runtime": 59.7909,
"eval_samples_per_second": 142.63,
"step": 70400
},
{
"epoch": 57.01,
"learning_rate": 8.099813183880438e-06,
"loss": 0.0006,
"step": 71200
},
{
"epoch": 57.01,
"eval_loss": 0.08382847905158997,
"eval_micro_f1": 0.9125881977831556,
"eval_runtime": 57.2712,
"eval_samples_per_second": 148.906,
"step": 71200
},
{
"epoch": 57.65,
"learning_rate": 8.078462770216173e-06,
"loss": 0.0007,
"step": 72000
},
{
"epoch": 57.65,
"eval_loss": 0.08181499689817429,
"eval_micro_f1": 0.9127402002109648,
"eval_runtime": 61.7245,
"eval_samples_per_second": 138.162,
"step": 72000
},
{
"epoch": 58.29,
"learning_rate": 8.05711235655191e-06,
"loss": 0.0006,
"step": 72800
},
{
"epoch": 58.29,
"eval_loss": 0.08484584093093872,
"eval_micro_f1": 0.9126798556150376,
"eval_runtime": 57.0734,
"eval_samples_per_second": 149.422,
"step": 72800
},
{
"epoch": 58.93,
"learning_rate": 8.035761942887644e-06,
"loss": 0.0006,
"step": 73600
},
{
"epoch": 58.93,
"eval_loss": 0.08549146354198456,
"eval_micro_f1": 0.9112660215372572,
"eval_runtime": 57.0411,
"eval_samples_per_second": 149.506,
"step": 73600
},
{
"epoch": 59.57,
"learning_rate": 8.01441152922338e-06,
"loss": 0.0006,
"step": 74400
},
{
"epoch": 59.57,
"eval_loss": 0.08189860731363297,
"eval_micro_f1": 0.9140181862290142,
"eval_runtime": 59.4438,
"eval_samples_per_second": 143.463,
"step": 74400
},
{
"epoch": 60.21,
"learning_rate": 7.993061115559114e-06,
"loss": 0.0005,
"step": 75200
},
{
"epoch": 60.21,
"eval_loss": 0.08556392043828964,
"eval_micro_f1": 0.9132947971878137,
"eval_runtime": 59.2786,
"eval_samples_per_second": 143.863,
"step": 75200
},
{
"epoch": 60.85,
"learning_rate": 7.97171070189485e-06,
"loss": 0.0007,
"step": 76000
},
{
"epoch": 60.85,
"eval_loss": 0.07878432422876358,
"eval_micro_f1": 0.9129439226355731,
"eval_runtime": 59.5274,
"eval_samples_per_second": 143.262,
"step": 76000
},
{
"epoch": 61.49,
"learning_rate": 7.950360288230585e-06,
"loss": 0.0005,
"step": 76800
},
{
"epoch": 61.49,
"eval_loss": 0.08688512444496155,
"eval_micro_f1": 0.9137193056499554,
"eval_runtime": 57.1631,
"eval_samples_per_second": 149.187,
"step": 76800
},
{
"epoch": 62.13,
"learning_rate": 7.92900987456632e-06,
"loss": 0.0005,
"step": 77600
},
{
"epoch": 62.13,
"eval_loss": 0.08153047412633896,
"eval_micro_f1": 0.9127534799601429,
"eval_runtime": 56.8967,
"eval_samples_per_second": 149.886,
"step": 77600
},
{
"epoch": 62.77,
"learning_rate": 7.907659460902055e-06,
"loss": 0.0005,
"step": 78400
},
{
"epoch": 62.77,
"eval_loss": 0.08603304624557495,
"eval_micro_f1": 0.9136181148862692,
"eval_runtime": 57.0765,
"eval_samples_per_second": 149.414,
"step": 78400
},
{
"epoch": 63.41,
"learning_rate": 7.886309047237791e-06,
"loss": 0.0006,
"step": 79200
},
{
"epoch": 63.41,
"eval_loss": 0.087012380361557,
"eval_micro_f1": 0.9107006421676777,
"eval_runtime": 63.9094,
"eval_samples_per_second": 133.439,
"step": 79200
},
{
"epoch": 64.05,
"learning_rate": 7.864958633573526e-06,
"loss": 0.0006,
"step": 80000
},
{
"epoch": 64.05,
"eval_loss": 0.08591558039188385,
"eval_micro_f1": 0.9125786158521357,
"eval_runtime": 57.3064,
"eval_samples_per_second": 148.814,
"step": 80000
},
{
"epoch": 64.69,
"learning_rate": 7.843608219909261e-06,
"loss": 0.0005,
"step": 80800
},
{
"epoch": 64.69,
"eval_loss": 0.08318696916103363,
"eval_micro_f1": 0.9124675552146335,
"eval_runtime": 56.8918,
"eval_samples_per_second": 149.899,
"step": 80800
},
{
"epoch": 65.33,
"learning_rate": 7.822257806244996e-06,
"loss": 0.0005,
"step": 81600
},
{
"epoch": 65.33,
"eval_loss": 0.08768671751022339,
"eval_micro_f1": 0.9134125631672133,
"eval_runtime": 57.0404,
"eval_samples_per_second": 149.508,
"step": 81600
},
{
"epoch": 65.97,
"learning_rate": 7.800907392580732e-06,
"loss": 0.0005,
"step": 82400
},
{
"epoch": 65.97,
"eval_loss": 0.08384938538074493,
"eval_micro_f1": 0.9106783070889455,
"eval_runtime": 57.1743,
"eval_samples_per_second": 149.158,
"step": 82400
},
{
"epoch": 66.61,
"learning_rate": 7.779556978916467e-06,
"loss": 0.0006,
"step": 83200
},
{
"epoch": 66.61,
"eval_loss": 0.08304194360971451,
"eval_micro_f1": 0.9097954397089875,
"eval_runtime": 56.9076,
"eval_samples_per_second": 149.857,
"step": 83200
},
{
"epoch": 67.25,
"learning_rate": 7.758206565252202e-06,
"loss": 0.0005,
"step": 84000
},
{
"epoch": 67.25,
"eval_loss": 0.08840597420930862,
"eval_micro_f1": 0.9127228846052202,
"eval_runtime": 59.2313,
"eval_samples_per_second": 143.978,
"step": 84000
},
{
"epoch": 67.89,
"learning_rate": 7.736856151587939e-06,
"loss": 0.0005,
"step": 84800
},
{
"epoch": 67.89,
"eval_loss": 0.08252264559268951,
"eval_micro_f1": 0.9154022776503898,
"eval_runtime": 61.8482,
"eval_samples_per_second": 137.886,
"step": 84800
},
{
"epoch": 68.53,
"learning_rate": 7.715505737923673e-06,
"loss": 0.0004,
"step": 85600
},
{
"epoch": 68.53,
"eval_loss": 0.08666601777076721,
"eval_micro_f1": 0.9122901512191244,
"eval_runtime": 59.4665,
"eval_samples_per_second": 143.408,
"step": 85600
},
{
"epoch": 69.18,
"learning_rate": 7.694155324259408e-06,
"loss": 0.0005,
"step": 86400
},
{
"epoch": 69.18,
"eval_loss": 0.08830396085977554,
"eval_micro_f1": 0.9128500152309758,
"eval_runtime": 59.114,
"eval_samples_per_second": 144.264,
"step": 86400
},
{
"epoch": 69.82,
"learning_rate": 7.672804910595143e-06,
"loss": 0.0005,
"step": 87200
},
{
"epoch": 69.82,
"eval_loss": 0.0914982408285141,
"eval_micro_f1": 0.910376302690265,
"eval_runtime": 58.9693,
"eval_samples_per_second": 144.618,
"step": 87200
},
{
"epoch": 70.46,
"learning_rate": 7.65145449693088e-06,
"loss": 0.0005,
"step": 88000
},
{
"epoch": 70.46,
"eval_loss": 0.08569258451461792,
"eval_micro_f1": 0.9125943221733096,
"eval_runtime": 57.0814,
"eval_samples_per_second": 149.401,
"step": 88000
},
{
"epoch": 71.1,
"learning_rate": 7.630104083266614e-06,
"loss": 0.0005,
"step": 88800
},
{
"epoch": 71.1,
"eval_loss": 0.08814612030982971,
"eval_micro_f1": 0.9143113498812799,
"eval_runtime": 57.0046,
"eval_samples_per_second": 149.602,
"step": 88800
},
{
"epoch": 71.74,
"learning_rate": 7.608753669602349e-06,
"loss": 0.0006,
"step": 89600
},
{
"epoch": 71.74,
"eval_loss": 0.08136966824531555,
"eval_micro_f1": 0.9140649561505793,
"eval_runtime": 59.2787,
"eval_samples_per_second": 143.863,
"step": 89600
},
{
"epoch": 72.38,
"learning_rate": 7.587403255938084e-06,
"loss": 0.0004,
"step": 90400
},
{
"epoch": 72.38,
"eval_loss": 0.083094023168087,
"eval_micro_f1": 0.9136114998841677,
"eval_runtime": 61.4262,
"eval_samples_per_second": 138.833,
"step": 90400
},
{
"epoch": 73.02,
"learning_rate": 7.56605284227382e-06,
"loss": 0.0005,
"step": 91200
},
{
"epoch": 73.02,
"eval_loss": 0.08627807348966599,
"eval_micro_f1": 0.9144864226998152,
"eval_runtime": 56.8517,
"eval_samples_per_second": 150.004,
"step": 91200
},
{
"epoch": 73.66,
"learning_rate": 7.544702428609555e-06,
"loss": 0.0004,
"step": 92000
},
{
"epoch": 73.66,
"eval_loss": 0.09016964584589005,
"eval_micro_f1": 0.9122585868136597,
"eval_runtime": 56.9093,
"eval_samples_per_second": 149.852,
"step": 92000
},
{
"epoch": 74.3,
"learning_rate": 7.52335201494529e-06,
"loss": 0.0004,
"step": 92800
},
{
"epoch": 74.3,
"eval_loss": 0.09021825343370438,
"eval_micro_f1": 0.9113674096332266,
"eval_runtime": 57.0312,
"eval_samples_per_second": 149.532,
"step": 92800
},
{
"epoch": 74.94,
"learning_rate": 7.502001601281026e-06,
"loss": 0.0006,
"step": 93600
},
{
"epoch": 74.94,
"eval_loss": 0.08460699021816254,
"eval_micro_f1": 0.912129379553843,
"eval_runtime": 59.5004,
"eval_samples_per_second": 143.327,
"step": 93600
},
{
"epoch": 75.58,
"learning_rate": 7.4806511876167606e-06,
"loss": 0.0005,
"step": 94400
},
{
"epoch": 75.58,
"eval_loss": 0.08845075219869614,
"eval_micro_f1": 0.9123552816996284,
"eval_runtime": 57.7997,
"eval_samples_per_second": 147.544,
"step": 94400
},
{
"epoch": 76.22,
"learning_rate": 7.459300773952496e-06,
"loss": 0.0005,
"step": 95200
},
{
"epoch": 76.22,
"eval_loss": 0.08736979961395264,
"eval_micro_f1": 0.9141985275918456,
"eval_runtime": 57.1528,
"eval_samples_per_second": 149.214,
"step": 95200
},
{
"epoch": 76.86,
"learning_rate": 7.437950360288231e-06,
"loss": 0.0005,
"step": 96000
},
{
"epoch": 76.86,
"eval_loss": 0.0876312181353569,
"eval_micro_f1": 0.9137970038892572,
"eval_runtime": 59.3693,
"eval_samples_per_second": 143.643,
"step": 96000
},
{
"epoch": 77.5,
"learning_rate": 7.416599946623967e-06,
"loss": 0.0004,
"step": 96800
},
{
"epoch": 77.5,
"eval_loss": 0.09083260595798492,
"eval_micro_f1": 0.9119465626628013,
"eval_runtime": 59.4056,
"eval_samples_per_second": 143.555,
"step": 96800
},
{
"epoch": 78.14,
"learning_rate": 7.3952495329597015e-06,
"loss": 0.0005,
"step": 97600
},
{
"epoch": 78.14,
"eval_loss": 0.09044567495584488,
"eval_micro_f1": 0.9127969757418269,
"eval_runtime": 61.521,
"eval_samples_per_second": 138.619,
"step": 97600
},
{
"epoch": 78.78,
"learning_rate": 7.373899119295437e-06,
"loss": 0.0004,
"step": 98400
},
{
"epoch": 78.78,
"eval_loss": 0.0885007306933403,
"eval_micro_f1": 0.9128140358537146,
"eval_runtime": 59.3351,
"eval_samples_per_second": 143.726,
"step": 98400
},
{
"epoch": 79.42,
"learning_rate": 7.352548705631172e-06,
"loss": 0.0005,
"step": 99200
},
{
"epoch": 79.42,
"eval_loss": 0.0865035280585289,
"eval_micro_f1": 0.9125916590437664,
"eval_runtime": 57.1701,
"eval_samples_per_second": 149.169,
"step": 99200
},
{
"epoch": 80.06,
"learning_rate": 7.331198291966908e-06,
"loss": 0.0005,
"step": 100000
},
{
"epoch": 80.06,
"eval_loss": 0.08545194566249847,
"eval_micro_f1": 0.9129139964316033,
"eval_runtime": 57.3276,
"eval_samples_per_second": 148.759,
"step": 100000
},
{
"epoch": 80.7,
"learning_rate": 7.3098478783026425e-06,
"loss": 0.0004,
"step": 100800
},
{
"epoch": 80.7,
"eval_loss": 0.08943796157836914,
"eval_micro_f1": 0.9145769433931177,
"eval_runtime": 56.9096,
"eval_samples_per_second": 149.852,
"step": 100800
},
{
"epoch": 81.35,
"learning_rate": 7.288497464638378e-06,
"loss": 0.0004,
"step": 101600
},
{
"epoch": 81.35,
"eval_loss": 0.08977091312408447,
"eval_micro_f1": 0.9142908529939483,
"eval_runtime": 57.1277,
"eval_samples_per_second": 149.28,
"step": 101600
},
{
"epoch": 81.99,
"learning_rate": 7.267147050974114e-06,
"loss": 0.0005,
"step": 102400
},
{
"epoch": 81.99,
"eval_loss": 0.08894480019807816,
"eval_micro_f1": 0.9134649905232792,
"eval_runtime": 57.2197,
"eval_samples_per_second": 149.039,
"step": 102400
},
{
"epoch": 82.63,
"learning_rate": 7.245796637309849e-06,
"loss": 0.0004,
"step": 103200
},
{
"epoch": 82.63,
"eval_loss": 0.08727879077196121,
"eval_micro_f1": 0.9145990645844334,
"eval_runtime": 63.2499,
"eval_samples_per_second": 134.83,
"step": 103200
},
{
"epoch": 83.27,
"learning_rate": 7.224446223645583e-06,
"loss": 0.0004,
"step": 104000
},
{
"epoch": 83.27,
"eval_loss": 0.08676949143409729,
"eval_micro_f1": 0.9123043102701411,
"eval_runtime": 56.8066,
"eval_samples_per_second": 150.124,
"step": 104000
},
{
"epoch": 83.91,
"learning_rate": 7.203095809981318e-06,
"loss": 0.0004,
"step": 104800
},
{
"epoch": 83.91,
"eval_loss": 0.09100856631994247,
"eval_micro_f1": 0.914119228545221,
"eval_runtime": 57.3104,
"eval_samples_per_second": 148.804,
"step": 104800
},
{
"epoch": 84.55,
"learning_rate": 7.181745396317055e-06,
"loss": 0.0005,
"step": 105600
},
{
"epoch": 84.55,
"eval_loss": 0.09105472266674042,
"eval_micro_f1": 0.9110732227590956,
"eval_runtime": 59.5719,
"eval_samples_per_second": 143.155,
"step": 105600
},
{
"epoch": 85.19,
"learning_rate": 7.1603949826527895e-06,
"loss": 0.0003,
"step": 106400
},
{
"epoch": 85.19,
"eval_loss": 0.09283676743507385,
"eval_micro_f1": 0.9121606406236915,
"eval_runtime": 56.954,
"eval_samples_per_second": 149.735,
"step": 106400
},
{
"epoch": 85.83,
"learning_rate": 7.139044568988524e-06,
"loss": 0.0005,
"step": 107200
},
{
"epoch": 85.83,
"eval_loss": 0.08533892035484314,
"eval_micro_f1": 0.9135314711709448,
"eval_runtime": 57.0761,
"eval_samples_per_second": 149.414,
"step": 107200
},
{
"epoch": 86.47,
"learning_rate": 7.117694155324259e-06,
"loss": 0.0004,
"step": 108000
},
{
"epoch": 86.47,
"eval_loss": 0.08738704770803452,
"eval_micro_f1": 0.9153180610949636,
"eval_runtime": 59.2595,
"eval_samples_per_second": 143.909,
"step": 108000
},
{
"epoch": 87.11,
"learning_rate": 7.096343741659996e-06,
"loss": 0.0003,
"step": 108800
},
{
"epoch": 87.11,
"eval_loss": 0.09139962494373322,
"eval_micro_f1": 0.9102858423988672,
"eval_runtime": 63.5723,
"eval_samples_per_second": 134.146,
"step": 108800
},
{
"epoch": 87.75,
"learning_rate": 7.0749933279957305e-06,
"loss": 0.0004,
"step": 109600
},
{
"epoch": 87.75,
"eval_loss": 0.08956313133239746,
"eval_micro_f1": 0.9129632126117252,
"eval_runtime": 56.9545,
"eval_samples_per_second": 149.734,
"step": 109600
},
{
"epoch": 88.39,
"learning_rate": 7.053642914331465e-06,
"loss": 0.0004,
"step": 110400
},
{
"epoch": 88.39,
"eval_loss": 0.09099085628986359,
"eval_micro_f1": 0.9119474200455265,
"eval_runtime": 56.9127,
"eval_samples_per_second": 149.844,
"step": 110400
},
{
"epoch": 89.03,
"learning_rate": 7.0322925006672e-06,
"loss": 0.0003,
"step": 111200
},
{
"epoch": 89.03,
"eval_loss": 0.09258583933115005,
"eval_micro_f1": 0.9112282119036523,
"eval_runtime": 59.4524,
"eval_samples_per_second": 143.442,
"step": 111200
},
{
"epoch": 89.67,
"learning_rate": 7.010942087002937e-06,
"loss": 0.0003,
"step": 112000
},
{
"epoch": 89.67,
"eval_loss": 0.09471511840820312,
"eval_micro_f1": 0.9128108273249216,
"eval_runtime": 57.226,
"eval_samples_per_second": 149.023,
"step": 112000
},
{
"epoch": 90.31,
"learning_rate": 6.9895916733386714e-06,
"loss": 0.0005,
"step": 112800
},
{
"epoch": 90.31,
"eval_loss": 0.08982954174280167,
"eval_micro_f1": 0.9130434777607886,
"eval_runtime": 57.2377,
"eval_samples_per_second": 148.993,
"step": 112800
},
{
"epoch": 90.95,
"learning_rate": 6.968241259674406e-06,
"loss": 0.0005,
"step": 113600
},
{
"epoch": 90.95,
"eval_loss": 0.08552589267492294,
"eval_micro_f1": 0.9137829907022647,
"eval_runtime": 59.396,
"eval_samples_per_second": 143.579,
"step": 113600
},
{
"epoch": 91.59,
"learning_rate": 6.946890846010143e-06,
"loss": 0.0004,
"step": 114400
},
{
"epoch": 91.59,
"eval_loss": 0.0870257243514061,
"eval_micro_f1": 0.913501060975142,
"eval_runtime": 59.1931,
"eval_samples_per_second": 144.071,
"step": 114400
},
{
"epoch": 92.23,
"learning_rate": 6.925540432345878e-06,
"loss": 0.0004,
"step": 115200
},
{
"epoch": 92.23,
"eval_loss": 0.08763734996318817,
"eval_micro_f1": 0.9122097978743587,
"eval_runtime": 57.0686,
"eval_samples_per_second": 149.434,
"step": 115200
},
{
"epoch": 92.87,
"learning_rate": 6.904190018681612e-06,
"loss": 0.0004,
"step": 116000
},
{
"epoch": 92.87,
"eval_loss": 0.0906001552939415,
"eval_micro_f1": 0.9111969106968496,
"eval_runtime": 57.3292,
"eval_samples_per_second": 148.755,
"step": 116000
},
{
"epoch": 93.51,
"learning_rate": 6.882839605017347e-06,
"loss": 0.0004,
"step": 116800
},
{
"epoch": 93.51,
"eval_loss": 0.09109827131032944,
"eval_micro_f1": 0.9126813536106514,
"eval_runtime": 57.0827,
"eval_samples_per_second": 149.397,
"step": 116800
},
{
"epoch": 94.16,
"learning_rate": 6.861489191353084e-06,
"loss": 0.0005,
"step": 117600
},
{
"epoch": 94.16,
"eval_loss": 0.08240609616041183,
"eval_micro_f1": 0.9118428753051387,
"eval_runtime": 59.7691,
"eval_samples_per_second": 142.682,
"step": 117600
},
{
"epoch": 94.8,
"learning_rate": 6.8401387776888185e-06,
"loss": 0.0003,
"step": 118400
},
{
"epoch": 94.8,
"eval_loss": 0.08634831756353378,
"eval_micro_f1": 0.9125129447023191,
"eval_runtime": 61.5625,
"eval_samples_per_second": 138.526,
"step": 118400
},
{
"epoch": 95.44,
"learning_rate": 6.818788364024553e-06,
"loss": 0.0003,
"step": 119200
},
{
"epoch": 95.44,
"eval_loss": 0.08992088586091995,
"eval_micro_f1": 0.9128835451665506,
"eval_runtime": 57.2255,
"eval_samples_per_second": 149.024,
"step": 119200
},
{
"epoch": 96.08,
"learning_rate": 6.797437950360288e-06,
"loss": 0.0004,
"step": 120000
},
{
"epoch": 96.08,
"eval_loss": 0.09084143489599228,
"eval_micro_f1": 0.9121670261377954,
"eval_runtime": 56.8211,
"eval_samples_per_second": 150.085,
"step": 120000
},
{
"epoch": 96.72,
"learning_rate": 6.776087536696025e-06,
"loss": 0.0003,
"step": 120800
},
{
"epoch": 96.72,
"eval_loss": 0.09444824606180191,
"eval_micro_f1": 0.9103868735753515,
"eval_runtime": 56.9861,
"eval_samples_per_second": 149.65,
"step": 120800
},
{
"epoch": 97.36,
"learning_rate": 6.7547371230317595e-06,
"loss": 0.0004,
"step": 121600
},
{
"epoch": 97.36,
"eval_loss": 0.09316682815551758,
"eval_micro_f1": 0.9107535070529553,
"eval_runtime": 56.8763,
"eval_samples_per_second": 149.939,
"step": 121600
},
{
"epoch": 98.0,
"learning_rate": 6.733386709367494e-06,
"loss": 0.0004,
"step": 122400
},
{
"epoch": 98.0,
"eval_loss": 0.09283734858036041,
"eval_micro_f1": 0.9126283224479024,
"eval_runtime": 59.6632,
"eval_samples_per_second": 142.936,
"step": 122400
},
{
"epoch": 98.64,
"learning_rate": 6.71203629570323e-06,
"loss": 0.0003,
"step": 123200
},
{
"epoch": 98.64,
"eval_loss": 0.08649221062660217,
"eval_micro_f1": 0.9130864636379873,
"eval_runtime": 59.8706,
"eval_samples_per_second": 142.44,
"step": 123200
},
{
"epoch": 99.28,
"learning_rate": 6.690685882038966e-06,
"loss": 0.0003,
"step": 124000
},
{
"epoch": 99.28,
"eval_loss": 0.09055854380130768,
"eval_micro_f1": 0.913199425611856,
"eval_runtime": 57.6864,
"eval_samples_per_second": 147.834,
"step": 124000
},
{
"epoch": 99.92,
"learning_rate": 6.6693354683747004e-06,
"loss": 0.0003,
"step": 124800
},
{
"epoch": 99.92,
"eval_loss": 0.09059716016054153,
"eval_micro_f1": 0.9138428257436125,
"eval_runtime": 57.0459,
"eval_samples_per_second": 149.494,
"step": 124800
},
{
"epoch": 100.56,
"learning_rate": 6.647985054710435e-06,
"loss": 0.0003,
"step": 125600
},
{
"epoch": 100.56,
"eval_loss": 0.09224581718444824,
"eval_micro_f1": 0.9149309455613784,
"eval_runtime": 59.5513,
"eval_samples_per_second": 143.204,
"step": 125600
},
{
"epoch": 101.2,
"learning_rate": 6.626634641046171e-06,
"loss": 0.0003,
"step": 126400
},
{
"epoch": 101.2,
"eval_loss": 0.09310884773731232,
"eval_micro_f1": 0.9121116609137015,
"eval_runtime": 59.4886,
"eval_samples_per_second": 143.355,
"step": 126400
},
{
"epoch": 101.84,
"learning_rate": 6.6052842273819066e-06,
"loss": 0.0004,
"step": 127200
},
{
"epoch": 101.84,
"eval_loss": 0.08837177604436874,
"eval_micro_f1": 0.9124792307182225,
"eval_runtime": 56.7786,
"eval_samples_per_second": 150.197,
"step": 127200
},
{
"epoch": 102.48,
"learning_rate": 6.583933813717641e-06,
"loss": 0.0003,
"step": 128000
},
{
"epoch": 102.48,
"eval_loss": 0.08561829477548599,
"eval_micro_f1": 0.9132082275764254,
"eval_runtime": 57.1447,
"eval_samples_per_second": 149.235,
"step": 128000
},
{
"epoch": 103.12,
"learning_rate": 6.562583400053376e-06,
"loss": 0.0004,
"step": 128800
},
{
"epoch": 103.12,
"eval_loss": 0.0907057672739029,
"eval_micro_f1": 0.9128324941081291,
"eval_runtime": 59.3298,
"eval_samples_per_second": 143.739,
"step": 128800
},
{
"epoch": 103.76,
"learning_rate": 6.541232986389112e-06,
"loss": 0.0003,
"step": 129600
},
{
"epoch": 103.76,
"eval_loss": 0.08664930611848831,
"eval_micro_f1": 0.9123565153675817,
"eval_runtime": 56.8604,
"eval_samples_per_second": 149.981,
"step": 129600
},
{
"epoch": 104.4,
"learning_rate": 6.5198825727248475e-06,
"loss": 0.0003,
"step": 130400
},
{
"epoch": 104.4,
"eval_loss": 0.09194556623697281,
"eval_micro_f1": 0.9118483407321456,
"eval_runtime": 57.8021,
"eval_samples_per_second": 147.538,
"step": 130400
},
{
"epoch": 105.04,
"learning_rate": 6.498532159060582e-06,
"loss": 0.0004,
"step": 131200
},
{
"epoch": 105.04,
"eval_loss": 0.08820647746324539,
"eval_micro_f1": 0.9143294462203133,
"eval_runtime": 61.1308,
"eval_samples_per_second": 139.504,
"step": 131200
},
{
"epoch": 105.68,
"learning_rate": 6.477181745396317e-06,
"loss": 0.0003,
"step": 132000
},
{
"epoch": 105.68,
"eval_loss": 0.09362075477838516,
"eval_micro_f1": 0.9122255640848616,
"eval_runtime": 57.2523,
"eval_samples_per_second": 148.955,
"step": 132000
},
{
"epoch": 106.33,
"learning_rate": 6.455831331732053e-06,
"loss": 0.0002,
"step": 132800
},
{
"epoch": 106.33,
"eval_loss": 0.09215810894966125,
"eval_micro_f1": 0.913590263191604,
"eval_runtime": 56.9967,
"eval_samples_per_second": 149.623,
"step": 132800
},
{
"epoch": 106.97,
"learning_rate": 6.4344809180677885e-06,
"loss": 0.0003,
"step": 133600
},
{
"epoch": 106.97,
"eval_loss": 0.09212013334035873,
"eval_micro_f1": 0.9096055345884416,
"eval_runtime": 59.4941,
"eval_samples_per_second": 143.342,
"step": 133600
},
{
"epoch": 107.61,
"learning_rate": 6.413130504403523e-06,
"loss": 0.0003,
"step": 134400
},
{
"epoch": 107.61,
"eval_loss": 0.08978000283241272,
"eval_micro_f1": 0.9103374104422837,
"eval_runtime": 57.0661,
"eval_samples_per_second": 149.441,
"step": 134400
},
{
"epoch": 108.25,
"learning_rate": 6.391780090739259e-06,
"loss": 0.0003,
"step": 135200
},
{
"epoch": 108.25,
"eval_loss": 0.09154769778251648,
"eval_micro_f1": 0.9094906883040702,
"eval_runtime": 57.0369,
"eval_samples_per_second": 149.517,
"step": 135200
},
{
"epoch": 108.89,
"learning_rate": 6.370429677074994e-06,
"loss": 0.0002,
"step": 136000
},
{
"epoch": 108.89,
"eval_loss": 0.09310750663280487,
"eval_micro_f1": 0.9124054730325141,
"eval_runtime": 56.9692,
"eval_samples_per_second": 149.695,
"step": 136000
},
{
"epoch": 109.53,
"learning_rate": 6.349079263410729e-06,
"loss": 0.0003,
"step": 136800
},
{
"epoch": 109.53,
"eval_loss": 0.09188534319400787,
"eval_micro_f1": 0.9117291409751688,
"eval_runtime": 61.2671,
"eval_samples_per_second": 139.194,
"step": 136800
},
{
"epoch": 110.17,
"learning_rate": 6.327728849746464e-06,
"loss": 0.0003,
"step": 137600
},
{
"epoch": 110.17,
"eval_loss": 0.0898001417517662,
"eval_micro_f1": 0.9115652364188669,
"eval_runtime": 57.036,
"eval_samples_per_second": 149.52,
"step": 137600
},
{
"epoch": 110.81,
"learning_rate": 6.3063784360822e-06,
"loss": 0.0002,
"step": 138400
},
{
"epoch": 110.81,
"eval_loss": 0.09399331361055374,
"eval_micro_f1": 0.9114824987115986,
"eval_runtime": 56.8683,
"eval_samples_per_second": 149.96,
"step": 138400
},
{
"epoch": 111.45,
"learning_rate": 6.285028022417935e-06,
"loss": 0.0003,
"step": 139200
},
{
"epoch": 111.45,
"eval_loss": 0.09055063873529434,
"eval_micro_f1": 0.9132776725406072,
"eval_runtime": 57.0641,
"eval_samples_per_second": 149.446,
"step": 139200
},
{
"epoch": 112.09,
"learning_rate": 6.2636776087536695e-06,
"loss": 0.0003,
"step": 140000
},
{
"epoch": 112.09,
"eval_loss": 0.09382639080286026,
"eval_micro_f1": 0.9126674727266407,
"eval_runtime": 57.139,
"eval_samples_per_second": 149.25,
"step": 140000
},
{
"epoch": 112.73,
"learning_rate": 6.242327195089405e-06,
"loss": 0.0003,
"step": 140800
},
{
"epoch": 112.73,
"eval_loss": 0.09113107621669769,
"eval_micro_f1": 0.9108518396729032,
"eval_runtime": 58.0604,
"eval_samples_per_second": 146.882,
"step": 140800
},
{
"epoch": 113.37,
"learning_rate": 6.220976781425141e-06,
"loss": 0.0003,
"step": 141600
},
{
"epoch": 113.37,
"eval_loss": 0.09415099024772644,
"eval_micro_f1": 0.9117130104512934,
"eval_runtime": 58.1448,
"eval_samples_per_second": 146.668,
"step": 141600
},
{
"epoch": 114.01,
"learning_rate": 6.199626367760876e-06,
"loss": 0.0003,
"step": 142400
},
{
"epoch": 114.01,
"eval_loss": 0.09359373152256012,
"eval_micro_f1": 0.9106482728747288,
"eval_runtime": 62.6263,
"eval_samples_per_second": 136.173,
"step": 142400
},
{
"epoch": 114.65,
"learning_rate": 6.1782759540966105e-06,
"loss": 0.0003,
"step": 143200
},
{
"epoch": 114.65,
"eval_loss": 0.09276324510574341,
"eval_micro_f1": 0.9113362882902918,
"eval_runtime": 55.7921,
"eval_samples_per_second": 152.853,
"step": 143200
},
{
"epoch": 115.29,
"learning_rate": 6.156925540432347e-06,
"loss": 0.0002,
"step": 144000
},
{
"epoch": 115.29,
"eval_loss": 0.09371680021286011,
"eval_micro_f1": 0.9146226197798658,
"eval_runtime": 56.0155,
"eval_samples_per_second": 152.244,
"step": 144000
},
{
"epoch": 115.93,
"learning_rate": 6.135575126768082e-06,
"loss": 0.0003,
"step": 144800
},
{
"epoch": 115.93,
"eval_loss": 0.08909646421670914,
"eval_micro_f1": 0.9140769468635923,
"eval_runtime": 74.0278,
"eval_samples_per_second": 115.2,
"step": 144800
},
{
"epoch": 116.57,
"learning_rate": 6.114224713103817e-06,
"loss": 0.0002,
"step": 145600
},
{
"epoch": 116.57,
"eval_loss": 0.09569381177425385,
"eval_micro_f1": 0.9139958698653222,
"eval_runtime": 80.3205,
"eval_samples_per_second": 106.175,
"step": 145600
},
{
"epoch": 117.21,
"learning_rate": 6.0928742994395514e-06,
"loss": 0.0003,
"step": 146400
},
{
"epoch": 117.21,
"eval_loss": 0.09639108180999756,
"eval_micro_f1": 0.9114745230841903,
"eval_runtime": 68.2166,
"eval_samples_per_second": 125.014,
"step": 146400
},
{
"epoch": 117.85,
"learning_rate": 6.071523885775288e-06,
"loss": 0.0002,
"step": 147200
},
{
"epoch": 117.85,
"eval_loss": 0.0933372899889946,
"eval_micro_f1": 0.9140831795439105,
"eval_runtime": 66.4412,
"eval_samples_per_second": 128.354,
"step": 147200
},
{
"epoch": 118.49,
"learning_rate": 6.050173472111023e-06,
"loss": 0.0004,
"step": 148000
},
{
"epoch": 118.49,
"eval_loss": 0.09086288511753082,
"eval_micro_f1": 0.9140698767426111,
"eval_runtime": 68.4374,
"eval_samples_per_second": 124.61,
"step": 148000
},
{
"epoch": 119.14,
"learning_rate": 6.0288230584467576e-06,
"loss": 0.0003,
"step": 148800
},
{
"epoch": 119.14,
"eval_loss": 0.09362834692001343,
"eval_micro_f1": 0.9138936222463804,
"eval_runtime": 65.3635,
"eval_samples_per_second": 130.47,
"step": 148800
},
{
"epoch": 119.78,
"learning_rate": 6.007472644782492e-06,
"loss": 0.0003,
"step": 149600
},
{
"epoch": 119.78,
"eval_loss": 0.0919169932603836,
"eval_micro_f1": 0.9133611218162964,
"eval_runtime": 67.85,
"eval_samples_per_second": 125.689,
"step": 149600
},
{
"epoch": 120.42,
"learning_rate": 5.986122231118229e-06,
"loss": 0.0003,
"step": 150400
},
{
"epoch": 120.42,
"eval_loss": 0.08913933485746384,
"eval_micro_f1": 0.9144296051476652,
"eval_runtime": 68.8774,
"eval_samples_per_second": 123.814,
"step": 150400
},
{
"epoch": 121.06,
"learning_rate": 5.964771817453964e-06,
"loss": 0.0003,
"step": 151200
},
{
"epoch": 121.06,
"eval_loss": 0.09314560890197754,
"eval_micro_f1": 0.913735719568292,
"eval_runtime": 65.998,
"eval_samples_per_second": 129.216,
"step": 151200
},
{
"epoch": 121.7,
"learning_rate": 5.9434214037896985e-06,
"loss": 0.0003,
"step": 152000
},
{
"epoch": 121.7,
"eval_loss": 0.08881153911352158,
"eval_micro_f1": 0.9144121760050435,
"eval_runtime": 55.5836,
"eval_samples_per_second": 153.427,
"step": 152000
},
{
"epoch": 122.34,
"learning_rate": 5.922070990125433e-06,
"loss": 0.0002,
"step": 152800
},
{
"epoch": 122.34,
"eval_loss": 0.09259207546710968,
"eval_micro_f1": 0.9120251736963795,
"eval_runtime": 54.8389,
"eval_samples_per_second": 155.51,
"step": 152800
},
{
"epoch": 122.98,
"learning_rate": 5.90072057646117e-06,
"loss": 0.0002,
"step": 153600
},
{
"epoch": 122.98,
"eval_loss": 0.09240709245204926,
"eval_micro_f1": 0.9144168461521939,
"eval_runtime": 54.9972,
"eval_samples_per_second": 155.062,
"step": 153600
},
{
"epoch": 123.62,
"learning_rate": 5.879370162796905e-06,
"loss": 0.0002,
"step": 154400
},
{
"epoch": 123.62,
"eval_loss": 0.09843878448009491,
"eval_micro_f1": 0.914642359519694,
"eval_runtime": 55.1957,
"eval_samples_per_second": 154.505,
"step": 154400
},
{
"epoch": 124.26,
"learning_rate": 5.8580197491326395e-06,
"loss": 0.0002,
"step": 155200
},
{
"epoch": 124.26,
"eval_loss": 0.09486906975507736,
"eval_micro_f1": 0.91233912289116,
"eval_runtime": 57.2154,
"eval_samples_per_second": 149.051,
"step": 155200
},
{
"epoch": 124.9,
"learning_rate": 5.836669335468376e-06,
"loss": 0.0002,
"step": 156000
},
{
"epoch": 124.9,
"eval_loss": 0.09536690264940262,
"eval_micro_f1": 0.9120187112667506,
"eval_runtime": 57.3982,
"eval_samples_per_second": 148.576,
"step": 156000
},
{
"epoch": 125.54,
"learning_rate": 5.815318921804111e-06,
"loss": 0.0002,
"step": 156800
},
{
"epoch": 125.54,
"eval_loss": 0.09243762493133545,
"eval_micro_f1": 0.9122554526143878,
"eval_runtime": 57.2249,
"eval_samples_per_second": 149.026,
"step": 156800
},
{
"epoch": 126.18,
"learning_rate": 5.793968508139846e-06,
"loss": 0.0003,
"step": 157600
},
{
"epoch": 126.18,
"eval_loss": 0.09130287170410156,
"eval_micro_f1": 0.9129711995718213,
"eval_runtime": 57.3345,
"eval_samples_per_second": 148.741,
"step": 157600
},
{
"epoch": 126.82,
"learning_rate": 5.77261809447558e-06,
"loss": 0.0003,
"step": 158400
},
{
"epoch": 126.82,
"eval_loss": 0.09325707703828812,
"eval_micro_f1": 0.914028371488214,
"eval_runtime": 57.348,
"eval_samples_per_second": 148.706,
"step": 158400
},
{
"epoch": 127.46,
"learning_rate": 5.751267680811317e-06,
"loss": 0.0002,
"step": 159200
},
{
"epoch": 127.46,
"eval_loss": 0.09333635121583939,
"eval_micro_f1": 0.9138147280432268,
"eval_runtime": 57.2291,
"eval_samples_per_second": 149.015,
"step": 159200
},
{
"epoch": 128.1,
"learning_rate": 5.729917267147052e-06,
"loss": 0.0002,
"step": 160000
},
{
"epoch": 128.1,
"eval_loss": 0.09427008777856827,
"eval_micro_f1": 0.9147293774731925,
"eval_runtime": 55.1707,
"eval_samples_per_second": 154.575,
"step": 160000
},
{
"epoch": 128.74,
"learning_rate": 5.7085668534827865e-06,
"loss": 0.0003,
"step": 160800
},
{
"epoch": 128.74,
"eval_loss": 0.0944739356637001,
"eval_micro_f1": 0.9140340623229227,
"eval_runtime": 59.2223,
"eval_samples_per_second": 144.0,
"step": 160800
},
{
"epoch": 129.38,
"learning_rate": 5.687216439818521e-06,
"loss": 0.0002,
"step": 161600
},
{
"epoch": 129.38,
"eval_loss": 0.09413941949605942,
"eval_micro_f1": 0.9139519204165809,
"eval_runtime": 57.4526,
"eval_samples_per_second": 148.435,
"step": 161600
},
{
"epoch": 130.02,
"learning_rate": 5.665866026154258e-06,
"loss": 0.0002,
"step": 162400
},
{
"epoch": 130.02,
"eval_loss": 0.09581708163022995,
"eval_micro_f1": 0.9132177432513304,
"eval_runtime": 57.5612,
"eval_samples_per_second": 148.155,
"step": 162400
},
{
"epoch": 130.66,
"learning_rate": 5.644515612489993e-06,
"loss": 0.0002,
"step": 163200
},
{
"epoch": 130.66,
"eval_loss": 0.09603932499885559,
"eval_micro_f1": 0.9121203972424982,
"eval_runtime": 57.7464,
"eval_samples_per_second": 147.68,
"step": 163200
},
{
"epoch": 131.31,
"learning_rate": 5.6231651988257275e-06,
"loss": 0.0002,
"step": 164000
},
{
"epoch": 131.31,
"eval_loss": 0.09799838811159134,
"eval_micro_f1": 0.9141832842859701,
"eval_runtime": 57.4777,
"eval_samples_per_second": 148.371,
"step": 164000
},
{
"epoch": 131.95,
"learning_rate": 5.601814785161463e-06,
"loss": 0.0002,
"step": 164800
},
{
"epoch": 131.95,
"eval_loss": 0.09619379043579102,
"eval_micro_f1": 0.9127927922927153,
"eval_runtime": 57.0383,
"eval_samples_per_second": 149.514,
"step": 164800
},
{
"epoch": 132.59,
"learning_rate": 5.580464371497199e-06,
"loss": 0.0003,
"step": 165600
},
{
"epoch": 132.59,
"eval_loss": 0.09315957129001617,
"eval_micro_f1": 0.9146800496881577,
"eval_runtime": 55.1426,
"eval_samples_per_second": 154.654,
"step": 165600
},
{
"epoch": 133.23,
"learning_rate": 5.559113957832934e-06,
"loss": 0.0002,
"step": 166400
},
{
"epoch": 133.23,
"eval_loss": 0.09479640424251556,
"eval_micro_f1": 0.915362474765264,
"eval_runtime": 61.3212,
"eval_samples_per_second": 139.071,
"step": 166400
},
{
"epoch": 133.87,
"learning_rate": 5.5377635441686684e-06,
"loss": 0.0002,
"step": 167200
},
{
"epoch": 133.87,
"eval_loss": 0.0950850173830986,
"eval_micro_f1": 0.9140515955559323,
"eval_runtime": 57.2972,
"eval_samples_per_second": 148.838,
"step": 167200
},
{
"epoch": 134.51,
"learning_rate": 5.516413130504404e-06,
"loss": 0.0003,
"step": 168000
},
{
"epoch": 134.51,
"eval_loss": 0.0887296125292778,
"eval_micro_f1": 0.9125935212955724,
"eval_runtime": 55.2279,
"eval_samples_per_second": 154.415,
"step": 168000
},
{
"epoch": 135.15,
"learning_rate": 5.49506271684014e-06,
"loss": 0.0002,
"step": 168800
},
{
"epoch": 135.15,
"eval_loss": 0.09395181387662888,
"eval_micro_f1": 0.9139596526427651,
"eval_runtime": 55.1868,
"eval_samples_per_second": 154.53,
"step": 168800
},
{
"epoch": 135.79,
"learning_rate": 5.473712303175875e-06,
"loss": 0.0002,
"step": 169600
},
{
"epoch": 135.79,
"eval_loss": 0.09703458845615387,
"eval_micro_f1": 0.914803029123937,
"eval_runtime": 57.3022,
"eval_samples_per_second": 148.825,
"step": 169600
},
{
"epoch": 136.43,
"learning_rate": 5.452361889511609e-06,
"loss": 0.0002,
"step": 170400
},
{
"epoch": 136.43,
"eval_loss": 0.09293199330568314,
"eval_micro_f1": 0.9127534920641392,
"eval_runtime": 55.0662,
"eval_samples_per_second": 154.868,
"step": 170400
},
{
"epoch": 137.07,
"learning_rate": 5.431011475847345e-06,
"loss": 0.0003,
"step": 171200
},
{
"epoch": 137.07,
"eval_loss": 0.0932011604309082,
"eval_micro_f1": 0.9153028735527652,
"eval_runtime": 57.403,
"eval_samples_per_second": 148.564,
"step": 171200
},
{
"epoch": 137.71,
"learning_rate": 5.40966106218308e-06,
"loss": 0.0002,
"step": 172000
},
{
"epoch": 137.71,
"eval_loss": 0.09387822449207306,
"eval_micro_f1": 0.9157913631771095,
"eval_runtime": 57.2385,
"eval_samples_per_second": 148.991,
"step": 172000
},
{
"epoch": 138.35,
"learning_rate": 5.3883106485188155e-06,
"loss": 0.0002,
"step": 172800
},
{
"epoch": 138.35,
"eval_loss": 0.09500721842050552,
"eval_micro_f1": 0.9148103140672268,
"eval_runtime": 57.3548,
"eval_samples_per_second": 148.689,
"step": 172800
},
{
"epoch": 138.99,
"learning_rate": 5.36696023485455e-06,
"loss": 0.0002,
"step": 173600
},
{
"epoch": 138.99,
"eval_loss": 0.09737177938222885,
"eval_micro_f1": 0.9157974837203825,
"eval_runtime": 57.3628,
"eval_samples_per_second": 148.668,
"step": 173600
},
{
"epoch": 139.63,
"learning_rate": 5.345609821190286e-06,
"loss": 0.0002,
"step": 174400
},
{
"epoch": 139.63,
"eval_loss": 0.09171438217163086,
"eval_micro_f1": 0.915146066915662,
"eval_runtime": 55.1746,
"eval_samples_per_second": 154.564,
"step": 174400
},
{
"epoch": 140.27,
"learning_rate": 5.324259407526021e-06,
"loss": 0.0002,
"step": 175200
},
{
"epoch": 140.27,
"eval_loss": 0.0950022041797638,
"eval_micro_f1": 0.9163264385505954,
"eval_runtime": 55.2728,
"eval_samples_per_second": 154.289,
"step": 175200
},
{
"epoch": 140.91,
"learning_rate": 5.3029089938617565e-06,
"loss": 0.0002,
"step": 176000
},
{
"epoch": 140.91,
"eval_loss": 0.09068471193313599,
"eval_micro_f1": 0.915209947238253,
"eval_runtime": 57.1845,
"eval_samples_per_second": 149.131,
"step": 176000
},
{
"epoch": 141.55,
"learning_rate": 5.281558580197492e-06,
"loss": 0.0002,
"step": 176800
},
{
"epoch": 141.55,
"eval_loss": 0.09527377784252167,
"eval_micro_f1": 0.9162786527950885,
"eval_runtime": 57.3726,
"eval_samples_per_second": 148.642,
"step": 176800
},
{
"epoch": 142.19,
"learning_rate": 5.260208166533227e-06,
"loss": 0.0002,
"step": 177600
},
{
"epoch": 142.19,
"eval_loss": 0.09539304673671722,
"eval_micro_f1": 0.9163157418169507,
"eval_runtime": 57.5543,
"eval_samples_per_second": 148.173,
"step": 177600
},
{
"epoch": 142.83,
"learning_rate": 5.238857752868962e-06,
"loss": 0.0003,
"step": 178400
},
{
"epoch": 142.83,
"eval_loss": 0.09300664812326431,
"eval_micro_f1": 0.9169026106185333,
"eval_runtime": 57.4079,
"eval_samples_per_second": 148.551,
"step": 178400
},
{
"epoch": 143.47,
"learning_rate": 5.2175073392046974e-06,
"loss": 0.0002,
"step": 179200
},
{
"epoch": 143.47,
"eval_loss": 0.09496041387319565,
"eval_micro_f1": 0.9159610875921584,
"eval_runtime": 58.6722,
"eval_samples_per_second": 145.35,
"step": 179200
},
{
"epoch": 144.12,
"learning_rate": 5.196156925540433e-06,
"loss": 0.0002,
"step": 180000
},
{
"epoch": 144.12,
"eval_loss": 0.09904171526432037,
"eval_micro_f1": 0.9158097969944367,
"eval_runtime": 57.458,
"eval_samples_per_second": 148.421,
"step": 180000
},
{
"epoch": 144.76,
"learning_rate": 5.174806511876168e-06,
"loss": 0.0003,
"step": 180800
},
{
"epoch": 144.76,
"eval_loss": 0.09724908322095871,
"eval_micro_f1": 0.9153088625259238,
"eval_runtime": 57.2395,
"eval_samples_per_second": 148.988,
"step": 180800
},
{
"epoch": 145.4,
"learning_rate": 5.153456098211903e-06,
"loss": 0.0002,
"step": 181600
},
{
"epoch": 145.4,
"eval_loss": 0.09198299050331116,
"eval_micro_f1": 0.9144401887317306,
"eval_runtime": 57.1887,
"eval_samples_per_second": 149.12,
"step": 181600
},
{
"epoch": 146.04,
"learning_rate": 5.132105684547638e-06,
"loss": 0.0002,
"step": 182400
},
{
"epoch": 146.04,
"eval_loss": 0.09802019596099854,
"eval_micro_f1": 0.9141318972119215,
"eval_runtime": 55.2085,
"eval_samples_per_second": 154.469,
"step": 182400
},
{
"epoch": 146.68,
"learning_rate": 5.110755270883374e-06,
"loss": 0.0002,
"step": 183200
},
{
"epoch": 146.68,
"eval_loss": 0.09556955099105835,
"eval_micro_f1": 0.9158685187678818,
"eval_runtime": 55.2515,
"eval_samples_per_second": 154.349,
"step": 183200
},
{
"epoch": 147.32,
"learning_rate": 5.089404857219109e-06,
"loss": 0.0002,
"step": 184000
},
{
"epoch": 147.32,
"eval_loss": 0.0990467295050621,
"eval_micro_f1": 0.9154910528194319,
"eval_runtime": 57.4502,
"eval_samples_per_second": 148.442,
"step": 184000
},
{
"epoch": 147.96,
"learning_rate": 5.068054443554844e-06,
"loss": 0.0002,
"step": 184800
},
{
"epoch": 147.96,
"eval_loss": 0.09897967427968979,
"eval_micro_f1": 0.9145279865099771,
"eval_runtime": 54.9955,
"eval_samples_per_second": 155.067,
"step": 184800
},
{
"epoch": 148.6,
"learning_rate": 5.04670402989058e-06,
"loss": 0.0002,
"step": 185600
},
{
"epoch": 148.6,
"eval_loss": 0.09991764277219772,
"eval_micro_f1": 0.9120136882607306,
"eval_runtime": 57.3421,
"eval_samples_per_second": 148.721,
"step": 185600
},
{
"epoch": 149.24,
"learning_rate": 5.025353616226315e-06,
"loss": 0.0002,
"step": 186400
},
{
"epoch": 149.24,
"eval_loss": 0.10025490820407867,
"eval_micro_f1": 0.9118358565020152,
"eval_runtime": 57.1674,
"eval_samples_per_second": 149.176,
"step": 186400
},
{
"epoch": 149.88,
"learning_rate": 5.00400320256205e-06,
"loss": 0.0002,
"step": 187200
},
{
"epoch": 149.88,
"eval_loss": 0.09600796550512314,
"eval_micro_f1": 0.9146987514079924,
"eval_runtime": 57.1711,
"eval_samples_per_second": 149.166,
"step": 187200
},
{
"epoch": 150.52,
"learning_rate": 4.9826527888977855e-06,
"loss": 0.0002,
"step": 188000
},
{
"epoch": 150.52,
"eval_loss": 0.0957193449139595,
"eval_micro_f1": 0.9149431457082156,
"eval_runtime": 54.9708,
"eval_samples_per_second": 155.137,
"step": 188000
},
{
"epoch": 151.16,
"learning_rate": 4.96130237523352e-06,
"loss": 0.0002,
"step": 188800
},
{
"epoch": 151.16,
"eval_loss": 0.09552114456892014,
"eval_micro_f1": 0.9121815567299416,
"eval_runtime": 57.2077,
"eval_samples_per_second": 149.071,
"step": 188800
},
{
"epoch": 151.8,
"learning_rate": 4.939951961569256e-06,
"loss": 0.0002,
"step": 189600
},
{
"epoch": 151.8,
"eval_loss": 0.09797852486371994,
"eval_micro_f1": 0.9104946579073371,
"eval_runtime": 55.3116,
"eval_samples_per_second": 154.181,
"step": 189600
},
{
"epoch": 152.44,
"learning_rate": 4.918601547904991e-06,
"loss": 0.0002,
"step": 190400
},
{
"epoch": 152.44,
"eval_loss": 0.0995921939611435,
"eval_micro_f1": 0.9129458264696592,
"eval_runtime": 54.998,
"eval_samples_per_second": 155.06,
"step": 190400
},
{
"epoch": 153.08,
"learning_rate": 4.8972511342407264e-06,
"loss": 0.0003,
"step": 191200
},
{
"epoch": 153.08,
"eval_loss": 0.09367653727531433,
"eval_micro_f1": 0.9139770420387497,
"eval_runtime": 57.2252,
"eval_samples_per_second": 149.025,
"step": 191200
},
{
"epoch": 153.72,
"learning_rate": 4.875900720576461e-06,
"loss": 0.0002,
"step": 192000
},
{
"epoch": 153.72,
"eval_loss": 0.09356961399316788,
"eval_micro_f1": 0.9144807756407906,
"eval_runtime": 61.2963,
"eval_samples_per_second": 139.128,
"step": 192000
},
{
"epoch": 154.36,
"learning_rate": 4.854550306912197e-06,
"loss": 0.0002,
"step": 192800
},
{
"epoch": 154.36,
"eval_loss": 0.09716017544269562,
"eval_micro_f1": 0.9137028581766548,
"eval_runtime": 54.9805,
"eval_samples_per_second": 155.109,
"step": 192800
},
{
"epoch": 155.0,
"learning_rate": 4.833199893247932e-06,
"loss": 0.0002,
"step": 193600
},
{
"epoch": 155.0,
"eval_loss": 0.09624577313661575,
"eval_micro_f1": 0.9136771801325613,
"eval_runtime": 57.2914,
"eval_samples_per_second": 148.853,
"step": 193600
},
{
"epoch": 155.64,
"learning_rate": 4.811849479583667e-06,
"loss": 0.0003,
"step": 194400
},
{
"epoch": 155.64,
"eval_loss": 0.09600042551755905,
"eval_micro_f1": 0.9108572451597287,
"eval_runtime": 55.1096,
"eval_samples_per_second": 154.746,
"step": 194400
},
{
"epoch": 156.29,
"learning_rate": 4.790499065919403e-06,
"loss": 0.0002,
"step": 195200
},
{
"epoch": 156.29,
"eval_loss": 0.09662088006734848,
"eval_micro_f1": 0.9133872411890629,
"eval_runtime": 57.2287,
"eval_samples_per_second": 149.016,
"step": 195200
},
{
"epoch": 156.93,
"learning_rate": 4.769148652255138e-06,
"loss": 0.0002,
"step": 196000
},
{
"epoch": 156.93,
"eval_loss": 0.09260782599449158,
"eval_micro_f1": 0.9132728989722505,
"eval_runtime": 57.3868,
"eval_samples_per_second": 148.606,
"step": 196000
},
{
"epoch": 157.57,
"learning_rate": 4.7477982385908735e-06,
"loss": 0.0002,
"step": 196800
},
{
"epoch": 157.57,
"eval_loss": 0.09577851742506027,
"eval_micro_f1": 0.9120977884627223,
"eval_runtime": 55.0049,
"eval_samples_per_second": 155.041,
"step": 196800
},
{
"epoch": 158.21,
"learning_rate": 4.726447824926608e-06,
"loss": 0.0002,
"step": 197600
},
{
"epoch": 158.21,
"eval_loss": 0.09467930346727371,
"eval_micro_f1": 0.9152847058052664,
"eval_runtime": 58.9031,
"eval_samples_per_second": 144.78,
"step": 197600
},
{
"epoch": 158.85,
"learning_rate": 4.705097411262344e-06,
"loss": 0.0002,
"step": 198400
},
{
"epoch": 158.85,
"eval_loss": 0.09578932821750641,
"eval_micro_f1": 0.9141954407095771,
"eval_runtime": 57.3607,
"eval_samples_per_second": 148.673,
"step": 198400
},
{
"epoch": 159.49,
"learning_rate": 4.683746997598079e-06,
"loss": 0.0001,
"step": 199200
},
{
"epoch": 159.49,
"eval_loss": 0.09749435633420944,
"eval_micro_f1": 0.9144243290448786,
"eval_runtime": 54.7822,
"eval_samples_per_second": 155.671,
"step": 199200
},
{
"epoch": 160.13,
"learning_rate": 4.6623965839338145e-06,
"loss": 0.0002,
"step": 200000
},
{
"epoch": 160.13,
"eval_loss": 0.09928904473781586,
"eval_micro_f1": 0.9147711943214717,
"eval_runtime": 57.2813,
"eval_samples_per_second": 148.879,
"step": 200000
},
{
"epoch": 160.77,
"learning_rate": 4.641046170269549e-06,
"loss": 0.0002,
"step": 200800
},
{
"epoch": 160.77,
"eval_loss": 0.10227934271097183,
"eval_micro_f1": 0.9150038281562891,
"eval_runtime": 54.8597,
"eval_samples_per_second": 155.451,
"step": 200800
},
{
"epoch": 161.41,
"learning_rate": 4.619695756605285e-06,
"loss": 0.0002,
"step": 201600
},
{
"epoch": 161.41,
"eval_loss": 0.098277747631073,
"eval_micro_f1": 0.9141341480343667,
"eval_runtime": 54.8918,
"eval_samples_per_second": 155.36,
"step": 201600
},
{
"epoch": 162.05,
"learning_rate": 4.59834534294102e-06,
"loss": 0.0002,
"step": 202400
},
{
"epoch": 162.05,
"eval_loss": 0.0939130112528801,
"eval_micro_f1": 0.9153919041973595,
"eval_runtime": 57.3356,
"eval_samples_per_second": 148.738,
"step": 202400
},
{
"epoch": 162.69,
"learning_rate": 4.576994929276755e-06,
"loss": 0.0001,
"step": 203200
},
{
"epoch": 162.69,
"eval_loss": 0.09554446488618851,
"eval_micro_f1": 0.9152328044495628,
"eval_runtime": 61.6324,
"eval_samples_per_second": 138.369,
"step": 203200
},
{
"epoch": 163.33,
"learning_rate": 4.55564451561249e-06,
"loss": 0.0002,
"step": 204000
},
{
"epoch": 163.33,
"eval_loss": 0.09482933580875397,
"eval_micro_f1": 0.9131349290143964,
"eval_runtime": 54.8652,
"eval_samples_per_second": 155.435,
"step": 204000
},
{
"epoch": 163.97,
"learning_rate": 4.534294101948226e-06,
"loss": 0.0002,
"step": 204800
},
{
"epoch": 163.97,
"eval_loss": 0.09967362880706787,
"eval_micro_f1": 0.9127583550695438,
"eval_runtime": 57.4257,
"eval_samples_per_second": 148.505,
"step": 204800
},
{
"epoch": 164.61,
"learning_rate": 4.512943688283961e-06,
"loss": 0.0002,
"step": 205600
},
{
"epoch": 164.61,
"eval_loss": 0.09877758473157883,
"eval_micro_f1": 0.9123251703685198,
"eval_runtime": 57.2043,
"eval_samples_per_second": 149.08,
"step": 205600
},
{
"epoch": 165.25,
"learning_rate": 4.491593274619696e-06,
"loss": 0.0002,
"step": 206400
},
{
"epoch": 165.25,
"eval_loss": 0.0966004878282547,
"eval_micro_f1": 0.9138070786677156,
"eval_runtime": 57.4195,
"eval_samples_per_second": 148.521,
"step": 206400
},
{
"epoch": 165.89,
"learning_rate": 4.470242860955431e-06,
"loss": 0.0002,
"step": 207200
},
{
"epoch": 165.89,
"eval_loss": 0.09606374800205231,
"eval_micro_f1": 0.9144447431932692,
"eval_runtime": 55.0717,
"eval_samples_per_second": 154.853,
"step": 207200
},
{
"epoch": 166.53,
"learning_rate": 4.448892447291167e-06,
"loss": 0.0001,
"step": 208000
},
{
"epoch": 166.53,
"eval_loss": 0.10124710202217102,
"eval_micro_f1": 0.9141418677009558,
"eval_runtime": 57.1766,
"eval_samples_per_second": 149.152,
"step": 208000
},
{
"epoch": 167.17,
"learning_rate": 4.427542033626902e-06,
"loss": 0.0002,
"step": 208800
},
{
"epoch": 167.17,
"eval_loss": 0.09706932306289673,
"eval_micro_f1": 0.914133909176181,
"eval_runtime": 61.5164,
"eval_samples_per_second": 138.63,
"step": 208800
},
{
"epoch": 167.81,
"learning_rate": 4.406191619962637e-06,
"loss": 0.0002,
"step": 209600
},
{
"epoch": 167.81,
"eval_loss": 0.10038356482982635,
"eval_micro_f1": 0.9130473792558649,
"eval_runtime": 55.1359,
"eval_samples_per_second": 154.672,
"step": 209600
},
{
"epoch": 168.45,
"learning_rate": 4.384841206298372e-06,
"loss": 0.0002,
"step": 210400
},
{
"epoch": 168.45,
"eval_loss": 0.10104508697986603,
"eval_micro_f1": 0.9139601645815874,
"eval_runtime": 57.2698,
"eval_samples_per_second": 148.909,
"step": 210400
},
{
"epoch": 169.1,
"learning_rate": 4.363490792634108e-06,
"loss": 0.0002,
"step": 211200
},
{
"epoch": 169.1,
"eval_loss": 0.09737098962068558,
"eval_micro_f1": 0.914706409986561,
"eval_runtime": 55.0887,
"eval_samples_per_second": 154.805,
"step": 211200
},
{
"epoch": 169.74,
"learning_rate": 4.342140378969843e-06,
"loss": 0.0002,
"step": 212000
},
{
"epoch": 169.74,
"eval_loss": 0.0946039929986,
"eval_micro_f1": 0.9143804908035174,
"eval_runtime": 55.084,
"eval_samples_per_second": 154.818,
"step": 212000
},
{
"epoch": 170.38,
"learning_rate": 4.320789965305578e-06,
"loss": 0.0001,
"step": 212800
},
{
"epoch": 170.38,
"eval_loss": 0.09832293540239334,
"eval_micro_f1": 0.9139852592518647,
"eval_runtime": 57.2057,
"eval_samples_per_second": 149.076,
"step": 212800
},
{
"epoch": 171.02,
"learning_rate": 4.299439551641313e-06,
"loss": 0.0002,
"step": 213600
},
{
"epoch": 171.02,
"eval_loss": 0.0956304594874382,
"eval_micro_f1": 0.9148477494325685,
"eval_runtime": 55.1341,
"eval_samples_per_second": 154.677,
"step": 213600
},
{
"epoch": 171.66,
"learning_rate": 4.278089137977049e-06,
"loss": 0.0002,
"step": 214400
},
{
"epoch": 171.66,
"eval_loss": 0.09659715741872787,
"eval_micro_f1": 0.9163626531953691,
"eval_runtime": 57.441,
"eval_samples_per_second": 148.465,
"step": 214400
},
{
"epoch": 172.3,
"learning_rate": 4.2567387243127836e-06,
"loss": 0.0002,
"step": 215200
},
{
"epoch": 172.3,
"eval_loss": 0.0970970019698143,
"eval_micro_f1": 0.9159272200789598,
"eval_runtime": 54.9947,
"eval_samples_per_second": 155.069,
"step": 215200
},
{
"epoch": 172.94,
"learning_rate": 4.235388310648519e-06,
"loss": 0.0002,
"step": 216000
},
{
"epoch": 172.94,
"eval_loss": 0.09398230910301208,
"eval_micro_f1": 0.9156691319814544,
"eval_runtime": 61.5869,
"eval_samples_per_second": 138.471,
"step": 216000
},
{
"epoch": 173.58,
"learning_rate": 4.214037896984254e-06,
"loss": 0.0002,
"step": 216800
},
{
"epoch": 173.58,
"eval_loss": 0.09897588193416595,
"eval_micro_f1": 0.913623500375828,
"eval_runtime": 57.4285,
"eval_samples_per_second": 148.498,
"step": 216800
},
{
"epoch": 174.22,
"learning_rate": 4.19268748331999e-06,
"loss": 0.0002,
"step": 217600
},
{
"epoch": 174.22,
"eval_loss": 0.0963212102651596,
"eval_micro_f1": 0.9124842592588781,
"eval_runtime": 55.0803,
"eval_samples_per_second": 154.829,
"step": 217600
},
{
"epoch": 174.86,
"learning_rate": 4.1713370696557245e-06,
"loss": 0.0002,
"step": 218400
},
{
"epoch": 174.86,
"eval_loss": 0.09685526043176651,
"eval_micro_f1": 0.9136490245695716,
"eval_runtime": 55.1155,
"eval_samples_per_second": 154.73,
"step": 218400
},
{
"epoch": 175.5,
"learning_rate": 4.14998665599146e-06,
"loss": 0.0001,
"step": 219200
},
{
"epoch": 175.5,
"eval_loss": 0.09647747874259949,
"eval_micro_f1": 0.9142908208423686,
"eval_runtime": 54.9205,
"eval_samples_per_second": 155.279,
"step": 219200
},
{
"epoch": 176.14,
"learning_rate": 4.128636242327195e-06,
"loss": 0.0002,
"step": 220000
},
{
"epoch": 176.14,
"eval_loss": 0.0956219881772995,
"eval_micro_f1": 0.9135724863674456,
"eval_runtime": 57.1817,
"eval_samples_per_second": 149.139,
"step": 220000
},
{
"epoch": 176.78,
"learning_rate": 4.107285828662931e-06,
"loss": 0.0002,
"step": 220800
},
{
"epoch": 176.78,
"eval_loss": 0.0988043025135994,
"eval_micro_f1": 0.9125900724534606,
"eval_runtime": 57.3872,
"eval_samples_per_second": 148.605,
"step": 220800
},
{
"epoch": 177.42,
"learning_rate": 4.0859354149986655e-06,
"loss": 0.0001,
"step": 221600
},
{
"epoch": 177.42,
"eval_loss": 0.09940287470817566,
"eval_micro_f1": 0.9135436321053072,
"eval_runtime": 57.2073,
"eval_samples_per_second": 149.072,
"step": 221600
},
{
"epoch": 178.06,
"learning_rate": 4.064585001334401e-06,
"loss": 0.0002,
"step": 222400
},
{
"epoch": 178.06,
"eval_loss": 0.09859929978847504,
"eval_micro_f1": 0.9152557591442528,
"eval_runtime": 57.6341,
"eval_samples_per_second": 147.968,
"step": 222400
},
{
"epoch": 178.7,
"learning_rate": 4.043234587670136e-06,
"loss": 0.0002,
"step": 223200
},
{
"epoch": 178.7,
"eval_loss": 0.097477987408638,
"eval_micro_f1": 0.9135913182748895,
"eval_runtime": 61.0933,
"eval_samples_per_second": 139.59,
"step": 223200
},
{
"epoch": 179.34,
"learning_rate": 4.021884174005872e-06,
"loss": 0.0002,
"step": 224000
},
{
"epoch": 179.34,
"eval_loss": 0.09950955212116241,
"eval_micro_f1": 0.9156496279619809,
"eval_runtime": 57.2975,
"eval_samples_per_second": 148.837,
"step": 224000
},
{
"epoch": 179.98,
"learning_rate": 4.000533760341606e-06,
"loss": 0.0002,
"step": 224800
},
{
"epoch": 179.98,
"eval_loss": 0.09578191488981247,
"eval_micro_f1": 0.915194664243973,
"eval_runtime": 57.5504,
"eval_samples_per_second": 148.183,
"step": 224800
},
{
"epoch": 180.62,
"learning_rate": 3.979183346677342e-06,
"loss": 0.0002,
"step": 225600
},
{
"epoch": 180.62,
"eval_loss": 0.09787262231111526,
"eval_micro_f1": 0.9154853754195559,
"eval_runtime": 54.973,
"eval_samples_per_second": 155.131,
"step": 225600
},
{
"epoch": 181.27,
"learning_rate": 3.957832933013078e-06,
"loss": 0.0002,
"step": 226400
},
{
"epoch": 181.27,
"eval_loss": 0.09851614385843277,
"eval_micro_f1": 0.9156311898474839,
"eval_runtime": 57.0711,
"eval_samples_per_second": 149.428,
"step": 226400
},
{
"epoch": 181.91,
"learning_rate": 3.9364825193488125e-06,
"loss": 0.0001,
"step": 227200
},
{
"epoch": 181.91,
"eval_loss": 0.09799177199602127,
"eval_micro_f1": 0.9155911313972169,
"eval_runtime": 57.1434,
"eval_samples_per_second": 149.239,
"step": 227200
},
{
"epoch": 182.55,
"learning_rate": 3.915132105684548e-06,
"loss": 0.0002,
"step": 228000
},
{
"epoch": 182.55,
"eval_loss": 0.0962972342967987,
"eval_micro_f1": 0.9161278730631559,
"eval_runtime": 57.4558,
"eval_samples_per_second": 148.427,
"step": 228000
},
{
"epoch": 183.19,
"learning_rate": 3.893781692020283e-06,
"loss": 0.0002,
"step": 228800
},
{
"epoch": 183.19,
"eval_loss": 0.09887988865375519,
"eval_micro_f1": 0.9137001073747966,
"eval_runtime": 55.1953,
"eval_samples_per_second": 154.506,
"step": 228800
},
{
"epoch": 183.83,
"learning_rate": 3.872431278356019e-06,
"loss": 0.0002,
"step": 229600
},
{
"epoch": 183.83,
"eval_loss": 0.10015502572059631,
"eval_micro_f1": 0.9146917294994792,
"eval_runtime": 57.5985,
"eval_samples_per_second": 148.059,
"step": 229600
},
{
"epoch": 184.47,
"learning_rate": 3.8510808646917535e-06,
"loss": 0.0002,
"step": 230400
},
{
"epoch": 184.47,
"eval_loss": 0.09985687583684921,
"eval_micro_f1": 0.9131789240563312,
"eval_runtime": 61.5317,
"eval_samples_per_second": 138.595,
"step": 230400
},
{
"epoch": 185.11,
"learning_rate": 3.829730451027489e-06,
"loss": 0.0001,
"step": 231200
},
{
"epoch": 185.11,
"eval_loss": 0.09944012761116028,
"eval_micro_f1": 0.9140983012341418,
"eval_runtime": 55.2619,
"eval_samples_per_second": 154.32,
"step": 231200
},
{
"epoch": 185.75,
"learning_rate": 3.808380037363224e-06,
"loss": 0.0002,
"step": 232000
},
{
"epoch": 185.75,
"eval_loss": 0.10187701135873795,
"eval_micro_f1": 0.9170718525523279,
"eval_runtime": 55.0364,
"eval_samples_per_second": 154.952,
"step": 232000
},
{
"epoch": 186.39,
"learning_rate": 3.7870296236989596e-06,
"loss": 0.0002,
"step": 232800
},
{
"epoch": 186.39,
"eval_loss": 0.10252540558576584,
"eval_micro_f1": 0.9153911750464183,
"eval_runtime": 54.9488,
"eval_samples_per_second": 155.199,
"step": 232800
},
{
"epoch": 187.03,
"learning_rate": 3.7656792100346944e-06,
"loss": 0.0002,
"step": 233600
},
{
"epoch": 187.03,
"eval_loss": 0.09944195300340652,
"eval_micro_f1": 0.915712424844586,
"eval_runtime": 57.58,
"eval_samples_per_second": 148.107,
"step": 233600
},
{
"epoch": 187.67,
"learning_rate": 3.74432879637043e-06,
"loss": 0.0002,
"step": 234400
},
{
"epoch": 187.67,
"eval_loss": 0.09849774092435837,
"eval_micro_f1": 0.9162419371662396,
"eval_runtime": 61.4529,
"eval_samples_per_second": 138.773,
"step": 234400
},
{
"epoch": 188.31,
"learning_rate": 3.722978382706165e-06,
"loss": 0.0002,
"step": 235200
},
{
"epoch": 188.31,
"eval_loss": 0.09956146031618118,
"eval_micro_f1": 0.9147056166669923,
"eval_runtime": 57.607,
"eval_samples_per_second": 148.037,
"step": 235200
},
{
"epoch": 188.95,
"learning_rate": 3.7016279690419006e-06,
"loss": 0.0001,
"step": 236000
},
{
"epoch": 188.95,
"eval_loss": 0.10023823380470276,
"eval_micro_f1": 0.9158324724541069,
"eval_runtime": 57.3708,
"eval_samples_per_second": 148.647,
"step": 236000
},
{
"epoch": 189.59,
"learning_rate": 3.680277555377636e-06,
"loss": 0.0002,
"step": 236800
},
{
"epoch": 189.59,
"eval_loss": 0.09992596507072449,
"eval_micro_f1": 0.9142651291829205,
"eval_runtime": 57.47,
"eval_samples_per_second": 148.391,
"step": 236800
},
{
"epoch": 190.23,
"learning_rate": 3.658927141713371e-06,
"loss": 0.0001,
"step": 237600
},
{
"epoch": 190.23,
"eval_loss": 0.0997430607676506,
"eval_micro_f1": 0.9132719509933108,
"eval_runtime": 57.2961,
"eval_samples_per_second": 148.841,
"step": 237600
},
{
"epoch": 190.87,
"learning_rate": 3.6375767280491063e-06,
"loss": 0.0002,
"step": 238400
},
{
"epoch": 190.87,
"eval_loss": 0.10087636858224869,
"eval_micro_f1": 0.9138942367273799,
"eval_runtime": 57.4181,
"eval_samples_per_second": 148.525,
"step": 238400
},
{
"epoch": 191.51,
"learning_rate": 3.6162263143848415e-06,
"loss": 0.0001,
"step": 239200
},
{
"epoch": 191.51,
"eval_loss": 0.10214927047491074,
"eval_micro_f1": 0.9144807756407906,
"eval_runtime": 57.5253,
"eval_samples_per_second": 148.248,
"step": 239200
},
{
"epoch": 192.15,
"learning_rate": 3.5948759007205768e-06,
"loss": 0.0001,
"step": 240000
},
{
"epoch": 192.15,
"eval_loss": 0.10319621115922928,
"eval_micro_f1": 0.9142007831778041,
"eval_runtime": 57.5017,
"eval_samples_per_second": 148.309,
"step": 240000
},
{
"epoch": 192.79,
"learning_rate": 3.573525487056312e-06,
"loss": 0.0002,
"step": 240800
},
{
"epoch": 192.79,
"eval_loss": 0.1037900447845459,
"eval_micro_f1": 0.9139828409858085,
"eval_runtime": 55.1088,
"eval_samples_per_second": 154.748,
"step": 240800
},
{
"epoch": 193.43,
"learning_rate": 3.5521750733920472e-06,
"loss": 0.0001,
"step": 241600
},
{
"epoch": 193.43,
"eval_loss": 0.10554201155900955,
"eval_micro_f1": 0.9146511727647968,
"eval_runtime": 61.4965,
"eval_samples_per_second": 138.675,
"step": 241600
},
{
"epoch": 194.08,
"learning_rate": 3.5308246597277825e-06,
"loss": 0.0002,
"step": 242400
},
{
"epoch": 194.08,
"eval_loss": 0.10550332069396973,
"eval_micro_f1": 0.9134489744942897,
"eval_runtime": 57.4363,
"eval_samples_per_second": 148.478,
"step": 242400
},
{
"epoch": 194.72,
"learning_rate": 3.5094742460635177e-06,
"loss": 0.0001,
"step": 243200
},
{
"epoch": 194.72,
"eval_loss": 0.1031009778380394,
"eval_micro_f1": 0.9138024108729701,
"eval_runtime": 57.2972,
"eval_samples_per_second": 148.838,
"step": 243200
},
{
"epoch": 195.36,
"learning_rate": 3.488123832399253e-06,
"loss": 0.0001,
"step": 244000
},
{
"epoch": 195.36,
"eval_loss": 0.10292581468820572,
"eval_micro_f1": 0.9129553739314089,
"eval_runtime": 57.5285,
"eval_samples_per_second": 148.24,
"step": 244000
},
{
"epoch": 196.0,
"learning_rate": 3.466773418734988e-06,
"loss": 0.0001,
"step": 244800
},
{
"epoch": 196.0,
"eval_loss": 0.10285958647727966,
"eval_micro_f1": 0.9161325020773844,
"eval_runtime": 57.4505,
"eval_samples_per_second": 148.441,
"step": 244800
},
{
"epoch": 196.64,
"learning_rate": 3.445423005070724e-06,
"loss": 0.0001,
"step": 245600
},
{
"epoch": 196.64,
"eval_loss": 0.10466793924570084,
"eval_micro_f1": 0.915342724195006,
"eval_runtime": 55.313,
"eval_samples_per_second": 154.177,
"step": 245600
},
{
"epoch": 197.28,
"learning_rate": 3.4240725914064587e-06,
"loss": 0.0002,
"step": 246400
},
{
"epoch": 197.28,
"eval_loss": 0.10088597238063812,
"eval_micro_f1": 0.9155599420699352,
"eval_runtime": 57.4288,
"eval_samples_per_second": 148.497,
"step": 246400
},
{
"epoch": 197.92,
"learning_rate": 3.4027221777421943e-06,
"loss": 0.0001,
"step": 247200
},
{
"epoch": 197.92,
"eval_loss": 0.0996263325214386,
"eval_micro_f1": 0.9138489203632388,
"eval_runtime": 61.9284,
"eval_samples_per_second": 137.707,
"step": 247200
},
{
"epoch": 198.56,
"learning_rate": 3.381371764077929e-06,
"loss": 0.0002,
"step": 248000
},
{
"epoch": 198.56,
"eval_loss": 0.10186070948839188,
"eval_micro_f1": 0.9155264451506543,
"eval_runtime": 55.1418,
"eval_samples_per_second": 154.656,
"step": 248000
},
{
"epoch": 199.2,
"learning_rate": 3.360021350413665e-06,
"loss": 0.0001,
"step": 248800
},
{
"epoch": 199.2,
"eval_loss": 0.09957974404096603,
"eval_micro_f1": 0.9159164561077748,
"eval_runtime": 57.3826,
"eval_samples_per_second": 148.617,
"step": 248800
},
{
"epoch": 199.84,
"learning_rate": 3.3386709367493996e-06,
"loss": 0.0001,
"step": 249600
},
{
"epoch": 199.84,
"eval_loss": 0.0999671146273613,
"eval_micro_f1": 0.9149079777206883,
"eval_runtime": 57.5514,
"eval_samples_per_second": 148.18,
"step": 249600
},
{
"epoch": 200.48,
"learning_rate": 3.3173205230851353e-06,
"loss": 0.0001,
"step": 250400
},
{
"epoch": 200.48,
"eval_loss": 0.10089126229286194,
"eval_micro_f1": 0.9154278009820824,
"eval_runtime": 57.3999,
"eval_samples_per_second": 148.572,
"step": 250400
},
{
"epoch": 201.12,
"learning_rate": 3.29597010942087e-06,
"loss": 0.0001,
"step": 251200
},
{
"epoch": 201.12,
"eval_loss": 0.10179542750120163,
"eval_micro_f1": 0.9157384112857778,
"eval_runtime": 55.21,
"eval_samples_per_second": 154.465,
"step": 251200
},
{
"epoch": 201.76,
"learning_rate": 3.2746196957566058e-06,
"loss": 0.0001,
"step": 252000
},
{
"epoch": 201.76,
"eval_loss": 0.10053855180740356,
"eval_micro_f1": 0.9150973072440998,
"eval_runtime": 57.3025,
"eval_samples_per_second": 148.824,
"step": 252000
},
{
"epoch": 202.4,
"learning_rate": 3.2532692820923406e-06,
"loss": 0.0001,
"step": 252800
},
{
"epoch": 202.4,
"eval_loss": 0.10337552428245544,
"eval_micro_f1": 0.91536978018793,
"eval_runtime": 61.3834,
"eval_samples_per_second": 138.93,
"step": 252800
},
{
"epoch": 203.04,
"learning_rate": 3.2319188684280762e-06,
"loss": 0.0001,
"step": 253600
},
{
"epoch": 203.04,
"eval_loss": 0.10305693000555038,
"eval_micro_f1": 0.9147607274262417,
"eval_runtime": 57.3463,
"eval_samples_per_second": 148.71,
"step": 253600
},
{
"epoch": 203.68,
"learning_rate": 3.210568454763811e-06,
"loss": 0.0001,
"step": 254400
},
{
"epoch": 203.68,
"eval_loss": 0.10468693822622299,
"eval_micro_f1": 0.9149297785499737,
"eval_runtime": 54.798,
"eval_samples_per_second": 155.626,
"step": 254400
},
{
"epoch": 204.32,
"learning_rate": 3.1892180410995467e-06,
"loss": 0.0002,
"step": 255200
},
{
"epoch": 204.32,
"eval_loss": 0.10264534503221512,
"eval_micro_f1": 0.9131902249623262,
"eval_runtime": 55.1487,
"eval_samples_per_second": 154.636,
"step": 255200
},
{
"epoch": 204.96,
"learning_rate": 3.167867627435282e-06,
"loss": 0.0002,
"step": 256000
},
{
"epoch": 204.96,
"eval_loss": 0.103822261095047,
"eval_micro_f1": 0.9149844890022454,
"eval_runtime": 57.3374,
"eval_samples_per_second": 148.734,
"step": 256000
},
{
"epoch": 205.6,
"learning_rate": 3.146517213771017e-06,
"loss": 0.0001,
"step": 256800
},
{
"epoch": 205.6,
"eval_loss": 0.10407021641731262,
"eval_micro_f1": 0.9150820404079998,
"eval_runtime": 54.7953,
"eval_samples_per_second": 155.634,
"step": 256800
},
{
"epoch": 206.24,
"learning_rate": 3.1251668001067524e-06,
"loss": 0.0001,
"step": 257600
},
{
"epoch": 206.24,
"eval_loss": 0.10335775464773178,
"eval_micro_f1": 0.9138942367273799,
"eval_runtime": 57.2926,
"eval_samples_per_second": 148.85,
"step": 257600
},
{
"epoch": 206.89,
"learning_rate": 3.1038163864424877e-06,
"loss": 0.0001,
"step": 258400
},
{
"epoch": 206.89,
"eval_loss": 0.10438382625579834,
"eval_micro_f1": 0.9145643222768515,
"eval_runtime": 57.5427,
"eval_samples_per_second": 148.203,
"step": 258400
},
{
"epoch": 207.53,
"learning_rate": 3.082465972778223e-06,
"loss": 0.0002,
"step": 259200
},
{
"epoch": 207.53,
"eval_loss": 0.10584603250026703,
"eval_micro_f1": 0.9166367868789392,
"eval_runtime": 57.3014,
"eval_samples_per_second": 148.827,
"step": 259200
},
{
"epoch": 208.17,
"learning_rate": 3.061115559113958e-06,
"loss": 0.0001,
"step": 260000
},
{
"epoch": 208.17,
"eval_loss": 0.10414166003465652,
"eval_micro_f1": 0.9173013165862366,
"eval_runtime": 59.1343,
"eval_samples_per_second": 144.214,
"step": 260000
},
{
"epoch": 208.81,
"learning_rate": 3.0397651454496934e-06,
"loss": 0.0001,
"step": 260800
},
{
"epoch": 208.81,
"eval_loss": 0.10670571774244308,
"eval_micro_f1": 0.9164384171731444,
"eval_runtime": 55.1028,
"eval_samples_per_second": 154.765,
"step": 260800
},
{
"epoch": 209.45,
"learning_rate": 3.0184147317854286e-06,
"loss": 0.0001,
"step": 261600
},
{
"epoch": 209.45,
"eval_loss": 0.10558865964412689,
"eval_micro_f1": 0.9132129570697367,
"eval_runtime": 55.1453,
"eval_samples_per_second": 154.646,
"step": 261600
},
{
"epoch": 210.09,
"learning_rate": 2.997064318121164e-06,
"loss": 0.0001,
"step": 262400
},
{
"epoch": 210.09,
"eval_loss": 0.10179650038480759,
"eval_micro_f1": 0.91686503430793,
"eval_runtime": 55.0849,
"eval_samples_per_second": 154.815,
"step": 262400
},
{
"epoch": 210.73,
"learning_rate": 2.975713904456899e-06,
"loss": 0.0002,
"step": 263200
},
{
"epoch": 210.73,
"eval_loss": 0.10531292110681534,
"eval_micro_f1": 0.915629073082886,
"eval_runtime": 57.4616,
"eval_samples_per_second": 148.412,
"step": 263200
},
{
"epoch": 211.37,
"learning_rate": 2.9543634907926343e-06,
"loss": 0.0001,
"step": 264000
},
{
"epoch": 211.37,
"eval_loss": 0.10534530133008957,
"eval_micro_f1": 0.9164714357947494,
"eval_runtime": 57.4262,
"eval_samples_per_second": 148.504,
"step": 264000
},
{
"epoch": 212.01,
"learning_rate": 2.9330130771283696e-06,
"loss": 0.0001,
"step": 264800
},
{
"epoch": 212.01,
"eval_loss": 0.10199680179357529,
"eval_micro_f1": 0.91404668685656,
"eval_runtime": 57.5986,
"eval_samples_per_second": 148.059,
"step": 264800
},
{
"epoch": 212.65,
"learning_rate": 2.911662663464105e-06,
"loss": 0.0002,
"step": 265600
},
{
"epoch": 212.65,
"eval_loss": 0.10087160021066666,
"eval_micro_f1": 0.9175030374404265,
"eval_runtime": 61.2412,
"eval_samples_per_second": 139.253,
"step": 265600
},
{
"epoch": 213.29,
"learning_rate": 2.8903122497998405e-06,
"loss": 0.0001,
"step": 266400
},
{
"epoch": 213.29,
"eval_loss": 0.10460913926362991,
"eval_micro_f1": 0.9168462638677535,
"eval_runtime": 57.4529,
"eval_samples_per_second": 148.435,
"step": 266400
},
{
"epoch": 213.93,
"learning_rate": 2.8689618361355753e-06,
"loss": 0.0001,
"step": 267200
},
{
"epoch": 213.93,
"eval_loss": 0.10347094386816025,
"eval_micro_f1": 0.9176174642380533,
"eval_runtime": 57.624,
"eval_samples_per_second": 147.994,
"step": 267200
},
{
"epoch": 214.57,
"learning_rate": 2.847611422471311e-06,
"loss": 0.0001,
"step": 268000
},
{
"epoch": 214.57,
"eval_loss": 0.10344100743532181,
"eval_micro_f1": 0.9178119100361792,
"eval_runtime": 57.5414,
"eval_samples_per_second": 148.206,
"step": 268000
},
{
"epoch": 215.21,
"learning_rate": 2.8262610088070457e-06,
"loss": 0.0001,
"step": 268800
},
{
"epoch": 215.21,
"eval_loss": 0.10337857156991959,
"eval_micro_f1": 0.9164763307284042,
"eval_runtime": 57.4609,
"eval_samples_per_second": 148.414,
"step": 268800
},
{
"epoch": 215.85,
"learning_rate": 2.8049105951427814e-06,
"loss": 0.0001,
"step": 269600
},
{
"epoch": 215.85,
"eval_loss": 0.10240375995635986,
"eval_micro_f1": 0.9168086055062248,
"eval_runtime": 55.3914,
"eval_samples_per_second": 153.959,
"step": 269600
},
{
"epoch": 216.49,
"learning_rate": 2.7835601814785162e-06,
"loss": 0.0001,
"step": 270400
},
{
"epoch": 216.49,
"eval_loss": 0.10527974367141724,
"eval_micro_f1": 0.9155431297700191,
"eval_runtime": 57.6054,
"eval_samples_per_second": 148.042,
"step": 270400
},
{
"epoch": 217.13,
"learning_rate": 2.762209767814252e-06,
"loss": 0.0001,
"step": 271200
},
{
"epoch": 217.13,
"eval_loss": 0.10640925914049149,
"eval_micro_f1": 0.9166180753016937,
"eval_runtime": 61.4875,
"eval_samples_per_second": 138.695,
"step": 271200
},
{
"epoch": 217.77,
"learning_rate": 2.7408593541499867e-06,
"loss": 0.0002,
"step": 272000
},
{
"epoch": 217.77,
"eval_loss": 0.10601688921451569,
"eval_micro_f1": 0.9168311496883519,
"eval_runtime": 57.4979,
"eval_samples_per_second": 148.318,
"step": 272000
},
{
"epoch": 218.41,
"learning_rate": 2.7195089404857224e-06,
"loss": 0.0001,
"step": 272800
},
{
"epoch": 218.41,
"eval_loss": 0.10229593515396118,
"eval_micro_f1": 0.9172587457900627,
"eval_runtime": 57.4494,
"eval_samples_per_second": 148.444,
"step": 272800
},
{
"epoch": 219.06,
"learning_rate": 2.698158526821457e-06,
"loss": 0.0001,
"step": 273600
},
{
"epoch": 219.06,
"eval_loss": 0.10352972894906998,
"eval_micro_f1": 0.9173230113366413,
"eval_runtime": 57.3939,
"eval_samples_per_second": 148.587,
"step": 273600
},
{
"epoch": 219.7,
"learning_rate": 2.676808113157193e-06,
"loss": 0.0001,
"step": 274400
},
{
"epoch": 219.7,
"eval_loss": 0.10362675040960312,
"eval_micro_f1": 0.9169216916691422,
"eval_runtime": 55.158,
"eval_samples_per_second": 154.611,
"step": 274400
},
{
"epoch": 220.34,
"learning_rate": 2.6554576994929276e-06,
"loss": 0.0002,
"step": 275200
},
{
"epoch": 220.34,
"eval_loss": 0.10175172239542007,
"eval_micro_f1": 0.9176607762512236,
"eval_runtime": 57.417,
"eval_samples_per_second": 148.527,
"step": 275200
},
{
"epoch": 220.98,
"learning_rate": 2.6341072858286633e-06,
"loss": 0.0001,
"step": 276000
},
{
"epoch": 220.98,
"eval_loss": 0.10186651349067688,
"eval_micro_f1": 0.9179017330847735,
"eval_runtime": 57.2077,
"eval_samples_per_second": 149.071,
"step": 276000
},
{
"epoch": 221.62,
"learning_rate": 2.6127568721643985e-06,
"loss": 0.0001,
"step": 276800
},
{
"epoch": 221.62,
"eval_loss": 0.10417181253433228,
"eval_micro_f1": 0.9175549853317992,
"eval_runtime": 59.2542,
"eval_samples_per_second": 143.922,
"step": 276800
},
{
"epoch": 222.26,
"learning_rate": 2.5914064585001338e-06,
"loss": 0.0001,
"step": 277600
},
{
"epoch": 222.26,
"eval_loss": 0.10326965898275375,
"eval_micro_f1": 0.9170415686919368,
"eval_runtime": 55.1512,
"eval_samples_per_second": 154.63,
"step": 277600
},
{
"epoch": 222.9,
"learning_rate": 2.570056044835869e-06,
"loss": 0.0002,
"step": 278400
},
{
"epoch": 222.9,
"eval_loss": 0.10621996223926544,
"eval_micro_f1": 0.9131977183018588,
"eval_runtime": 55.2608,
"eval_samples_per_second": 154.323,
"step": 278400
},
{
"epoch": 223.54,
"learning_rate": 2.5487056311716043e-06,
"loss": 0.0001,
"step": 279200
},
{
"epoch": 223.54,
"eval_loss": 0.10489922016859055,
"eval_micro_f1": 0.9155467656467245,
"eval_runtime": 57.4006,
"eval_samples_per_second": 148.57,
"step": 279200
},
{
"epoch": 224.18,
"learning_rate": 2.5273552175073395e-06,
"loss": 0.0001,
"step": 280000
},
{
"epoch": 224.18,
"eval_loss": 0.10579479485750198,
"eval_micro_f1": 0.9145456997566904,
"eval_runtime": 57.2904,
"eval_samples_per_second": 148.856,
"step": 280000
},
{
"epoch": 224.82,
"learning_rate": 2.5060048038430747e-06,
"loss": 0.0001,
"step": 280800
},
{
"epoch": 224.82,
"eval_loss": 0.10392695665359497,
"eval_micro_f1": 0.9171091705916365,
"eval_runtime": 55.408,
"eval_samples_per_second": 153.913,
"step": 280800
},
{
"epoch": 225.46,
"learning_rate": 2.48465439017881e-06,
"loss": 0.0001,
"step": 281600
},
{
"epoch": 225.46,
"eval_loss": 0.10539467632770538,
"eval_micro_f1": 0.9168391678916096,
"eval_runtime": 57.2217,
"eval_samples_per_second": 149.034,
"step": 281600
},
{
"epoch": 226.1,
"learning_rate": 2.463303976514545e-06,
"loss": 0.0001,
"step": 282400
},
{
"epoch": 226.1,
"eval_loss": 0.1050795242190361,
"eval_micro_f1": 0.916569480077678,
"eval_runtime": 55.3672,
"eval_samples_per_second": 154.026,
"step": 282400
},
{
"epoch": 226.74,
"learning_rate": 2.4419535628502804e-06,
"loss": 0.0001,
"step": 283200
},
{
"epoch": 226.74,
"eval_loss": 0.1055595725774765,
"eval_micro_f1": 0.9169928460379209,
"eval_runtime": 57.5097,
"eval_samples_per_second": 148.288,
"step": 283200
},
{
"epoch": 227.38,
"learning_rate": 2.4206031491860157e-06,
"loss": 0.0001,
"step": 284000
},
{
"epoch": 227.38,
"eval_loss": 0.10756581276655197,
"eval_micro_f1": 0.9169846894833256,
"eval_runtime": 55.0128,
"eval_samples_per_second": 155.018,
"step": 284000
},
{
"epoch": 228.02,
"learning_rate": 2.399252735521751e-06,
"loss": 0.0001,
"step": 284800
},
{
"epoch": 228.02,
"eval_loss": 0.10315944999456406,
"eval_micro_f1": 0.9164164383472595,
"eval_runtime": 55.2184,
"eval_samples_per_second": 154.441,
"step": 284800
},
{
"epoch": 228.66,
"learning_rate": 2.377902321857486e-06,
"loss": 0.0001,
"step": 285600
},
{
"epoch": 228.66,
"eval_loss": 0.10239244252443314,
"eval_micro_f1": 0.9158306711450752,
"eval_runtime": 57.2982,
"eval_samples_per_second": 148.835,
"step": 285600
},
{
"epoch": 229.3,
"learning_rate": 2.3565519081932214e-06,
"loss": 0.0001,
"step": 286400
},
{
"epoch": 229.3,
"eval_loss": 0.10360091179609299,
"eval_micro_f1": 0.9163642874540824,
"eval_runtime": 55.1046,
"eval_samples_per_second": 154.76,
"step": 286400
},
{
"epoch": 229.94,
"learning_rate": 2.3352014945289566e-06,
"loss": 0.0001,
"step": 287200
},
{
"epoch": 229.94,
"eval_loss": 0.10332570970058441,
"eval_micro_f1": 0.9168618261978149,
"eval_runtime": 55.1299,
"eval_samples_per_second": 154.689,
"step": 287200
},
{
"epoch": 230.58,
"learning_rate": 2.313851080864692e-06,
"loss": 0.0001,
"step": 288000
},
{
"epoch": 230.58,
"eval_loss": 0.10376840084791183,
"eval_micro_f1": 0.9184872813625782,
"eval_runtime": 59.2749,
"eval_samples_per_second": 143.872,
"step": 288000
}
],
"max_steps": 374700,
"num_train_epochs": 300,
"total_flos": 1059924613496689680,
"trial_name": null,
"trial_params": null
}
{
"best_metric": 0.9184872813625782,
"best_model_checkpoint": "../out/onto_final/ner-roberta-qa4-querytype4-span-char-pattern3-subtext-dim50/checkpoints/checkpoint-288000",
"epoch": 230.58446757405926,
"global_step": 288000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.64,
"learning_rate": 9.978649586335736e-06,
"loss": 0.0794,
"step": 800
},
{
"epoch": 0.64,
"eval_loss": 0.03532102704048157,
"eval_micro_f1": 0.8791638865358467,
"eval_runtime": 57.079,
"eval_samples_per_second": 149.407,
"step": 800
},
{
"epoch": 1.28,
"learning_rate": 9.957299172671472e-06,
"loss": 0.0358,
"step": 1600
},
{
"epoch": 1.28,
"eval_loss": 0.03331479802727699,
"eval_micro_f1": 0.8892428863535353,
"eval_runtime": 61.3385,
"eval_samples_per_second": 139.032,
"step": 1600
},
{
"epoch": 1.92,
"learning_rate": 9.935948759007207e-06,
"loss": 0.0309,
"step": 2400
},
{
"epoch": 1.92,
"eval_loss": 0.030986160039901733,
"eval_micro_f1": 0.8977142347839291,
"eval_runtime": 56.7577,
"eval_samples_per_second": 150.253,
"step": 2400
},
{
"epoch": 2.56,
"learning_rate": 9.914598345342942e-06,
"loss": 0.0265,
"step": 3200
},
{
"epoch": 2.56,
"eval_loss": 0.03181663900613785,
"eval_micro_f1": 0.9019206824095267,
"eval_runtime": 57.1062,
"eval_samples_per_second": 149.336,
"step": 3200
},
{
"epoch": 3.2,
"learning_rate": 9.893247931678677e-06,
"loss": 0.0238,
"step": 4000
},
{
"epoch": 3.2,
"eval_loss": 0.0325116403400898,
"eval_micro_f1": 0.9057191136154928,
"eval_runtime": 57.5307,
"eval_samples_per_second": 148.234,
"step": 4000
},
{
"epoch": 3.84,
"learning_rate": 9.871897518014413e-06,
"loss": 0.0217,
"step": 4800
},
{
"epoch": 3.84,
"eval_loss": 0.03316809609532356,
"eval_micro_f1": 0.9071629899701918,
"eval_runtime": 57.2457,
"eval_samples_per_second": 148.972,
"step": 4800
},
{
"epoch": 4.48,
"learning_rate": 9.850547104350148e-06,
"loss": 0.0179,
"step": 5600
},
{