diffcse-roberta-base-sts / trainer_state.json
voidism's picture
first commit
f6c6052
{
"best_metric": 0.8442339359569462,
"best_model_checkpoint": "result/simcse-celectra-amlp-dmlp-2ep-bs64-lr1e-5-mask0.20-elew0.005-roberta-base",
"epoch": 2.0,
"global_step": 31252,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"electra_acc": 0.1088,
"electra_fix_acc": 0.0079,
"electra_rep_acc": 0.9968,
"epoch": 0.0,
"learning_rate": 9.99968002047869e-06,
"loss": 8.4419,
"neg_sim": 0.2061,
"pos_sim": 0.3449,
"step": 1
},
{
"epoch": 0.01,
"eval_avg_sts": 0.755574714462574,
"eval_sickr_spearman": 0.7098302262826256,
"eval_stsb_spearman": 0.8013192026425223,
"step": 125
},
{
"epoch": 0.02,
"eval_avg_sts": 0.761612717263195,
"eval_sickr_spearman": 0.7112582869905835,
"eval_stsb_spearman": 0.8119671475358067,
"step": 250
},
{
"epoch": 0.02,
"eval_avg_sts": 0.766754789307388,
"eval_sickr_spearman": 0.7152875680591253,
"eval_stsb_spearman": 0.8182220105556509,
"step": 375
},
{
"electra_acc": 0.8686,
"electra_fix_acc": 0.9713,
"electra_rep_acc": 0.0311,
"epoch": 0.03,
"learning_rate": 9.840010239344683e-06,
"loss": 0.1729,
"neg_sim": -0.0093,
"pos_sim": 0.695,
"step": 500
},
{
"epoch": 0.03,
"eval_avg_sts": 0.7675239503020774,
"eval_sickr_spearman": 0.7166760511394983,
"eval_stsb_spearman": 0.8183718494646566,
"step": 500
},
{
"epoch": 0.04,
"eval_avg_sts": 0.7749845711206813,
"eval_sickr_spearman": 0.724027931681123,
"eval_stsb_spearman": 0.8259412105602397,
"step": 625
},
{
"epoch": 0.05,
"eval_avg_sts": 0.7751202672393465,
"eval_sickr_spearman": 0.7197742012755609,
"eval_stsb_spearman": 0.8304663332031321,
"step": 750
},
{
"epoch": 0.06,
"eval_avg_sts": 0.7760175671946854,
"eval_sickr_spearman": 0.7239823981969596,
"eval_stsb_spearman": 0.8280527361924113,
"step": 875
},
{
"electra_acc": 0.8964,
"electra_fix_acc": 0.9953,
"electra_rep_acc": 0.0862,
"epoch": 0.06,
"learning_rate": 9.680020478689364e-06,
"loss": 0.0014,
"neg_sim": -0.0136,
"pos_sim": 0.736,
"step": 1000
},
{
"epoch": 0.06,
"eval_avg_sts": 0.7775564387382596,
"eval_sickr_spearman": 0.7228041472475814,
"eval_stsb_spearman": 0.832308730228938,
"step": 1000
},
{
"epoch": 0.07,
"eval_avg_sts": 0.7784780411860817,
"eval_sickr_spearman": 0.7230463680921337,
"eval_stsb_spearman": 0.8339097142800296,
"step": 1125
},
{
"epoch": 0.08,
"eval_avg_sts": 0.7802639531360682,
"eval_sickr_spearman": 0.726064546444302,
"eval_stsb_spearman": 0.8344633598278346,
"step": 1250
},
{
"epoch": 0.09,
"eval_avg_sts": 0.778730071271978,
"eval_sickr_spearman": 0.7238748085297384,
"eval_stsb_spearman": 0.8335853340142177,
"step": 1375
},
{
"electra_acc": 0.8989,
"electra_fix_acc": 0.9938,
"electra_rep_acc": 0.1172,
"epoch": 0.1,
"learning_rate": 9.520030718034047e-06,
"loss": 0.0013,
"neg_sim": -0.0137,
"pos_sim": 0.7472,
"step": 1500
},
{
"epoch": 0.1,
"eval_avg_sts": 0.7716526802556551,
"eval_sickr_spearman": 0.7141475498664919,
"eval_stsb_spearman": 0.8291578106448184,
"step": 1500
},
{
"epoch": 0.1,
"eval_avg_sts": 0.7758744711124749,
"eval_sickr_spearman": 0.721181992858712,
"eval_stsb_spearman": 0.8305669493662378,
"step": 1625
},
{
"epoch": 0.11,
"eval_avg_sts": 0.7763031700578917,
"eval_sickr_spearman": 0.7214235412678442,
"eval_stsb_spearman": 0.8311827988479393,
"step": 1750
},
{
"epoch": 0.12,
"eval_avg_sts": 0.7806434242533656,
"eval_sickr_spearman": 0.7289663454376855,
"eval_stsb_spearman": 0.8323205030690458,
"step": 1875
},
{
"electra_acc": 0.8998,
"electra_fix_acc": 0.9931,
"electra_rep_acc": 0.1339,
"epoch": 0.13,
"learning_rate": 9.36004095737873e-06,
"loss": 0.0012,
"neg_sim": -0.0138,
"pos_sim": 0.7538,
"step": 2000
},
{
"epoch": 0.13,
"eval_avg_sts": 0.7795467000474285,
"eval_sickr_spearman": 0.722814521965492,
"eval_stsb_spearman": 0.836278878129365,
"step": 2000
},
{
"epoch": 0.14,
"eval_avg_sts": 0.7831655326956792,
"eval_sickr_spearman": 0.7220971294344124,
"eval_stsb_spearman": 0.8442339359569462,
"step": 2125
},
{
"epoch": 0.14,
"eval_avg_sts": 0.7784529836143297,
"eval_sickr_spearman": 0.7178746672758864,
"eval_stsb_spearman": 0.8390312999527729,
"step": 2250
},
{
"epoch": 0.15,
"eval_avg_sts": 0.7838110056478796,
"eval_sickr_spearman": 0.7265441850232628,
"eval_stsb_spearman": 0.8410778262724965,
"step": 2375
},
{
"electra_acc": 0.9008,
"electra_fix_acc": 0.9925,
"electra_rep_acc": 0.1492,
"epoch": 0.16,
"learning_rate": 9.20005119672341e-06,
"loss": 0.0012,
"neg_sim": -0.0138,
"pos_sim": 0.7534,
"step": 2500
},
{
"epoch": 0.16,
"eval_avg_sts": 0.781754791702129,
"eval_sickr_spearman": 0.723123794227652,
"eval_stsb_spearman": 0.8403857891766059,
"step": 2500
},
{
"epoch": 0.17,
"eval_avg_sts": 0.7798490649002168,
"eval_sickr_spearman": 0.7198115694724797,
"eval_stsb_spearman": 0.8398865603279537,
"step": 2625
},
{
"epoch": 0.18,
"eval_avg_sts": 0.7787504502510374,
"eval_sickr_spearman": 0.7188935990617945,
"eval_stsb_spearman": 0.8386073014402803,
"step": 2750
},
{
"epoch": 0.18,
"eval_avg_sts": 0.7702513371927302,
"eval_sickr_spearman": 0.7122062768396675,
"eval_stsb_spearman": 0.828296397545793,
"step": 2875
},
{
"electra_acc": 0.9012,
"electra_fix_acc": 0.9922,
"electra_rep_acc": 0.157,
"epoch": 0.19,
"learning_rate": 9.040061436068092e-06,
"loss": 0.0012,
"neg_sim": -0.0139,
"pos_sim": 0.7662,
"step": 3000
},
{
"epoch": 0.19,
"eval_avg_sts": 0.77796510106683,
"eval_sickr_spearman": 0.7215734943665341,
"eval_stsb_spearman": 0.834356707767126,
"step": 3000
},
{
"epoch": 0.2,
"eval_avg_sts": 0.7773533173320899,
"eval_sickr_spearman": 0.7216049067068746,
"eval_stsb_spearman": 0.8331017279573052,
"step": 3125
},
{
"epoch": 0.21,
"eval_avg_sts": 0.7727609962405579,
"eval_sickr_spearman": 0.7192614692677092,
"eval_stsb_spearman": 0.8262605232134068,
"step": 3250
},
{
"epoch": 0.22,
"eval_avg_sts": 0.7705397769228735,
"eval_sickr_spearman": 0.7183345170410549,
"eval_stsb_spearman": 0.8227450368046919,
"step": 3375
},
{
"electra_acc": 0.902,
"electra_fix_acc": 0.9917,
"electra_rep_acc": 0.1668,
"epoch": 0.22,
"learning_rate": 8.880071675412775e-06,
"loss": 0.0012,
"neg_sim": -0.0138,
"pos_sim": 0.762,
"step": 3500
},
{
"epoch": 0.22,
"eval_avg_sts": 0.7669127684681659,
"eval_sickr_spearman": 0.7153820452356542,
"eval_stsb_spearman": 0.8184434917006776,
"step": 3500
},
{
"epoch": 0.23,
"eval_avg_sts": 0.762067078104866,
"eval_sickr_spearman": 0.708602167081056,
"eval_stsb_spearman": 0.815531989128676,
"step": 3625
},
{
"epoch": 0.24,
"eval_avg_sts": 0.7603436095667805,
"eval_sickr_spearman": 0.7039441757894171,
"eval_stsb_spearman": 0.8167430433441438,
"step": 3750
},
{
"epoch": 0.25,
"eval_avg_sts": 0.7572330050419072,
"eval_sickr_spearman": 0.7035958853781614,
"eval_stsb_spearman": 0.810870124705653,
"step": 3875
},
{
"electra_acc": 0.903,
"electra_fix_acc": 0.9917,
"electra_rep_acc": 0.1735,
"epoch": 0.26,
"learning_rate": 8.720081914757458e-06,
"loss": 0.0013,
"neg_sim": -0.0139,
"pos_sim": 0.769,
"step": 4000
},
{
"epoch": 0.26,
"eval_avg_sts": 0.7565075332583117,
"eval_sickr_spearman": 0.7049569907307146,
"eval_stsb_spearman": 0.8080580757859087,
"step": 4000
},
{
"epoch": 0.26,
"eval_avg_sts": 0.7571155448037524,
"eval_sickr_spearman": 0.7046521373298869,
"eval_stsb_spearman": 0.809578952277618,
"step": 4125
},
{
"epoch": 0.27,
"eval_avg_sts": 0.7558024272832136,
"eval_sickr_spearman": 0.7044353249379953,
"eval_stsb_spearman": 0.8071695296284319,
"step": 4250
},
{
"epoch": 0.28,
"eval_avg_sts": 0.7590707229344114,
"eval_sickr_spearman": 0.7043194258902251,
"eval_stsb_spearman": 0.8138220199785977,
"step": 4375
},
{
"electra_acc": 0.9026,
"electra_fix_acc": 0.9912,
"electra_rep_acc": 0.1795,
"epoch": 0.29,
"learning_rate": 8.560092154102138e-06,
"loss": 0.0012,
"neg_sim": -0.0139,
"pos_sim": 0.7711,
"step": 4500
},
{
"epoch": 0.29,
"eval_avg_sts": 0.7618055470788606,
"eval_sickr_spearman": 0.7090844954016975,
"eval_stsb_spearman": 0.8145265987560237,
"step": 4500
},
{
"epoch": 0.3,
"eval_avg_sts": 0.7642277977124863,
"eval_sickr_spearman": 0.7120559394921663,
"eval_stsb_spearman": 0.8163996559328064,
"step": 4625
},
{
"epoch": 0.3,
"eval_avg_sts": 0.7631334021943239,
"eval_sickr_spearman": 0.7097783526930724,
"eval_stsb_spearman": 0.8164884516955754,
"step": 4750
},
{
"epoch": 0.31,
"eval_avg_sts": 0.7663792588185014,
"eval_sickr_spearman": 0.7141827086327445,
"eval_stsb_spearman": 0.8185758090042584,
"step": 4875
},
{
"electra_acc": 0.9031,
"electra_fix_acc": 0.9911,
"electra_rep_acc": 0.1837,
"epoch": 0.32,
"learning_rate": 8.400102393446819e-06,
"loss": 0.0011,
"neg_sim": -0.014,
"pos_sim": 0.7781,
"step": 5000
},
{
"epoch": 0.32,
"eval_avg_sts": 0.7675828244876146,
"eval_sickr_spearman": 0.7151267119004091,
"eval_stsb_spearman": 0.8200389370748202,
"step": 5000
},
{
"epoch": 0.33,
"eval_avg_sts": 0.7664182051626903,
"eval_sickr_spearman": 0.7138797284448728,
"eval_stsb_spearman": 0.8189566818805079,
"step": 5125
},
{
"epoch": 0.34,
"eval_avg_sts": 0.763298983477001,
"eval_sickr_spearman": 0.7031210499093442,
"eval_stsb_spearman": 0.8234769170446578,
"step": 5250
},
{
"epoch": 0.34,
"eval_avg_sts": 0.7609537736402838,
"eval_sickr_spearman": 0.7014618323453161,
"eval_stsb_spearman": 0.8204457149352516,
"step": 5375
},
{
"electra_acc": 0.9041,
"electra_fix_acc": 0.9909,
"electra_rep_acc": 0.1929,
"epoch": 0.35,
"learning_rate": 8.240112632791502e-06,
"loss": 0.0013,
"neg_sim": -0.0139,
"pos_sim": 0.7703,
"step": 5500
},
{
"epoch": 0.35,
"eval_avg_sts": 0.7649139683546902,
"eval_sickr_spearman": 0.7016405392386158,
"eval_stsb_spearman": 0.8281873974707646,
"step": 5500
},
{
"epoch": 0.36,
"eval_avg_sts": 0.7660619062755463,
"eval_sickr_spearman": 0.7030690322264868,
"eval_stsb_spearman": 0.8290547803246057,
"step": 5625
},
{
"epoch": 0.37,
"eval_avg_sts": 0.7667448016301607,
"eval_sickr_spearman": 0.7080724800943964,
"eval_stsb_spearman": 0.825417123165925,
"step": 5750
},
{
"epoch": 0.38,
"eval_avg_sts": 0.7668525663367054,
"eval_sickr_spearman": 0.7080771871423374,
"eval_stsb_spearman": 0.8256279455310735,
"step": 5875
},
{
"electra_acc": 0.9044,
"electra_fix_acc": 0.991,
"electra_rep_acc": 0.1933,
"epoch": 0.38,
"learning_rate": 8.080122872136184e-06,
"loss": 0.0011,
"neg_sim": -0.014,
"pos_sim": 0.7798,
"step": 6000
},
{
"epoch": 0.38,
"eval_avg_sts": 0.7666156811168598,
"eval_sickr_spearman": 0.709114370746792,
"eval_stsb_spearman": 0.8241169914869277,
"step": 6000
},
{
"epoch": 0.39,
"eval_avg_sts": 0.7644923521403739,
"eval_sickr_spearman": 0.7072277090823029,
"eval_stsb_spearman": 0.821756995198445,
"step": 6125
},
{
"epoch": 0.4,
"eval_avg_sts": 0.7661750461200847,
"eval_sickr_spearman": 0.7058679005694886,
"eval_stsb_spearman": 0.8264821916706808,
"step": 6250
},
{
"epoch": 0.41,
"eval_avg_sts": 0.7660352662732546,
"eval_sickr_spearman": 0.7060020514358052,
"eval_stsb_spearman": 0.8260684811107039,
"step": 6375
},
{
"electra_acc": 0.9042,
"electra_fix_acc": 0.9906,
"electra_rep_acc": 0.2009,
"epoch": 0.42,
"learning_rate": 7.920133111480865e-06,
"loss": 0.0011,
"neg_sim": -0.014,
"pos_sim": 0.7785,
"step": 6500
},
{
"epoch": 0.42,
"eval_avg_sts": 0.7705073293876892,
"eval_sickr_spearman": 0.7120254877738545,
"eval_stsb_spearman": 0.8289891710015237,
"step": 6500
},
{
"epoch": 0.42,
"eval_avg_sts": 0.7701345372028097,
"eval_sickr_spearman": 0.7109871514229654,
"eval_stsb_spearman": 0.829281922982654,
"step": 6625
},
{
"epoch": 0.43,
"eval_avg_sts": 0.7673565342593036,
"eval_sickr_spearman": 0.7094820007971993,
"eval_stsb_spearman": 0.8252310677214081,
"step": 6750
},
{
"epoch": 0.44,
"eval_avg_sts": 0.7663843153446501,
"eval_sickr_spearman": 0.7069766985461874,
"eval_stsb_spearman": 0.8257919321431128,
"step": 6875
},
{
"electra_acc": 0.9052,
"electra_fix_acc": 0.9907,
"electra_rep_acc": 0.2032,
"epoch": 0.45,
"learning_rate": 7.760143350825547e-06,
"loss": 0.0011,
"neg_sim": -0.0141,
"pos_sim": 0.7899,
"step": 7000
},
{
"epoch": 0.45,
"eval_avg_sts": 0.7671374698909221,
"eval_sickr_spearman": 0.7035898814904816,
"eval_stsb_spearman": 0.8306850582913625,
"step": 7000
},
{
"epoch": 0.46,
"eval_avg_sts": 0.7657223525531744,
"eval_sickr_spearman": 0.7024468853695589,
"eval_stsb_spearman": 0.82899781973679,
"step": 7125
},
{
"epoch": 0.46,
"eval_avg_sts": 0.7644676880553354,
"eval_sickr_spearman": 0.707080829974105,
"eval_stsb_spearman": 0.8218545461365657,
"step": 7250
},
{
"epoch": 0.47,
"eval_avg_sts": 0.7659127144064692,
"eval_sickr_spearman": 0.7085741649489175,
"eval_stsb_spearman": 0.8232512638640211,
"step": 7375
},
{
"electra_acc": 0.9051,
"electra_fix_acc": 0.9904,
"electra_rep_acc": 0.2082,
"epoch": 0.48,
"learning_rate": 7.600153590170229e-06,
"loss": 0.0011,
"neg_sim": -0.0141,
"pos_sim": 0.7901,
"step": 7500
},
{
"epoch": 0.48,
"eval_avg_sts": 0.7608800819642288,
"eval_sickr_spearman": 0.7044842686303607,
"eval_stsb_spearman": 0.817275895298097,
"step": 7500
},
{
"epoch": 0.49,
"eval_avg_sts": 0.7596082078600037,
"eval_sickr_spearman": 0.714175648060833,
"eval_stsb_spearman": 0.8050407676591743,
"step": 7625
},
{
"epoch": 0.5,
"eval_avg_sts": 0.7601954211728121,
"eval_sickr_spearman": 0.713427179407123,
"eval_stsb_spearman": 0.8069636629385014,
"step": 7750
},
{
"epoch": 0.5,
"eval_avg_sts": 0.7595124397232403,
"eval_sickr_spearman": 0.7107700028133637,
"eval_stsb_spearman": 0.808254876633117,
"step": 7875
},
{
"electra_acc": 0.9053,
"electra_fix_acc": 0.9904,
"electra_rep_acc": 0.2114,
"epoch": 0.51,
"learning_rate": 7.440163829514912e-06,
"loss": 0.0011,
"neg_sim": -0.014,
"pos_sim": 0.784,
"step": 8000
},
{
"epoch": 0.51,
"eval_avg_sts": 0.7600307296458793,
"eval_sickr_spearman": 0.7110970465830557,
"eval_stsb_spearman": 0.8089644127087027,
"step": 8000
},
{
"epoch": 0.52,
"eval_avg_sts": 0.7591940023385597,
"eval_sickr_spearman": 0.709230365856765,
"eval_stsb_spearman": 0.8091576388203543,
"step": 8125
},
{
"epoch": 0.53,
"eval_avg_sts": 0.7591314547260735,
"eval_sickr_spearman": 0.7092257548710269,
"eval_stsb_spearman": 0.8090371545811201,
"step": 8250
},
{
"epoch": 0.54,
"eval_avg_sts": 0.7574056509175848,
"eval_sickr_spearman": 0.7065820269856703,
"eval_stsb_spearman": 0.8082292748494992,
"step": 8375
},
{
"electra_acc": 0.9051,
"electra_fix_acc": 0.9903,
"electra_rep_acc": 0.2115,
"epoch": 0.54,
"learning_rate": 7.280174068859593e-06,
"loss": 0.001,
"neg_sim": -0.0141,
"pos_sim": 0.7991,
"step": 8500
},
{
"epoch": 0.54,
"eval_avg_sts": 0.7579907562127961,
"eval_sickr_spearman": 0.7058790918161235,
"eval_stsb_spearman": 0.8101024206094687,
"step": 8500
},
{
"epoch": 0.55,
"eval_avg_sts": 0.7588600922765898,
"eval_sickr_spearman": 0.7066013835195499,
"eval_stsb_spearman": 0.8111188010336298,
"step": 8625
},
{
"epoch": 0.56,
"eval_avg_sts": 0.7579781185788305,
"eval_sickr_spearman": 0.7017153716946564,
"eval_stsb_spearman": 0.8142408654630046,
"step": 8750
},
{
"epoch": 0.57,
"eval_avg_sts": 0.7583650057103912,
"eval_sickr_spearman": 0.7015226228845852,
"eval_stsb_spearman": 0.8152073885361973,
"step": 8875
},
{
"electra_acc": 0.9061,
"electra_fix_acc": 0.9906,
"electra_rep_acc": 0.2147,
"epoch": 0.58,
"learning_rate": 7.120184308204276e-06,
"loss": 0.0011,
"neg_sim": -0.0142,
"pos_sim": 0.8009,
"step": 9000
},
{
"epoch": 0.58,
"eval_avg_sts": 0.7584798299886761,
"eval_sickr_spearman": 0.7013053301816793,
"eval_stsb_spearman": 0.815654329795673,
"step": 9000
},
{
"epoch": 0.58,
"eval_avg_sts": 0.7568426056984886,
"eval_sickr_spearman": 0.6953886669821268,
"eval_stsb_spearman": 0.8182965444148504,
"step": 9125
},
{
"epoch": 0.59,
"eval_avg_sts": 0.7582279310008209,
"eval_sickr_spearman": 0.6970180260362119,
"eval_stsb_spearman": 0.8194378359654299,
"step": 9250
},
{
"epoch": 0.6,
"eval_avg_sts": 0.7573982613352882,
"eval_sickr_spearman": 0.6927659286880986,
"eval_stsb_spearman": 0.8220305939824778,
"step": 9375
},
{
"electra_acc": 0.9064,
"electra_fix_acc": 0.9904,
"electra_rep_acc": 0.2163,
"epoch": 0.61,
"learning_rate": 6.9601945475489575e-06,
"loss": 0.0011,
"neg_sim": -0.014,
"pos_sim": 0.7876,
"step": 9500
},
{
"epoch": 0.61,
"eval_avg_sts": 0.7553820601323997,
"eval_sickr_spearman": 0.690547148530363,
"eval_stsb_spearman": 0.8202169717344365,
"step": 9500
},
{
"epoch": 0.62,
"eval_avg_sts": 0.7567629669074094,
"eval_sickr_spearman": 0.6929571405029238,
"eval_stsb_spearman": 0.8205687933118949,
"step": 9625
},
{
"epoch": 0.62,
"eval_avg_sts": 0.7568625521954009,
"eval_sickr_spearman": 0.6927340360367437,
"eval_stsb_spearman": 0.8209910683540581,
"step": 9750
},
{
"epoch": 0.63,
"eval_avg_sts": 0.7524242981805003,
"eval_sickr_spearman": 0.6836492413863334,
"eval_stsb_spearman": 0.8211993549746672,
"step": 9875
},
{
"electra_acc": 0.9058,
"electra_fix_acc": 0.9902,
"electra_rep_acc": 0.2192,
"epoch": 0.64,
"learning_rate": 6.800204786893639e-06,
"loss": 0.001,
"neg_sim": -0.0141,
"pos_sim": 0.7942,
"step": 10000
},
{
"epoch": 0.64,
"eval_avg_sts": 0.7508138636055113,
"eval_sickr_spearman": 0.6821391435571191,
"eval_stsb_spearman": 0.8194885836539034,
"step": 10000
},
{
"epoch": 0.65,
"eval_avg_sts": 0.7510417193281674,
"eval_sickr_spearman": 0.683803901532964,
"eval_stsb_spearman": 0.8182795371233709,
"step": 10125
},
{
"epoch": 0.66,
"eval_avg_sts": 0.7560510909697331,
"eval_sickr_spearman": 0.694022134115111,
"eval_stsb_spearman": 0.8180800478243552,
"step": 10250
},
{
"epoch": 0.66,
"eval_avg_sts": 0.7554822011048945,
"eval_sickr_spearman": 0.6934548868071269,
"eval_stsb_spearman": 0.8175095154026621,
"step": 10375
},
{
"electra_acc": 0.9073,
"electra_fix_acc": 0.9904,
"electra_rep_acc": 0.2211,
"epoch": 0.67,
"learning_rate": 6.640215026238322e-06,
"loss": 0.0011,
"neg_sim": -0.0141,
"pos_sim": 0.7946,
"step": 10500
},
{
"epoch": 0.67,
"eval_avg_sts": 0.7614561025331773,
"eval_sickr_spearman": 0.7013578281755511,
"eval_stsb_spearman": 0.8215543768908035,
"step": 10500
},
{
"epoch": 0.68,
"eval_avg_sts": 0.7569156258584822,
"eval_sickr_spearman": 0.6959233972344373,
"eval_stsb_spearman": 0.817907854482527,
"step": 10625
},
{
"epoch": 0.69,
"eval_avg_sts": 0.7561592629720358,
"eval_sickr_spearman": 0.695029538436674,
"eval_stsb_spearman": 0.8172889875073976,
"step": 10750
},
{
"epoch": 0.7,
"eval_avg_sts": 0.7563428139477524,
"eval_sickr_spearman": 0.6957568734057513,
"eval_stsb_spearman": 0.8169287544897534,
"step": 10875
},
{
"electra_acc": 0.9065,
"electra_fix_acc": 0.9902,
"electra_rep_acc": 0.2217,
"epoch": 0.7,
"learning_rate": 6.480225265583003e-06,
"loss": 0.001,
"neg_sim": -0.0142,
"pos_sim": 0.8004,
"step": 11000
},
{
"epoch": 0.7,
"eval_avg_sts": 0.7516024333799518,
"eval_sickr_spearman": 0.6886655615894285,
"eval_stsb_spearman": 0.8145393051704751,
"step": 11000
},
{
"epoch": 0.71,
"eval_avg_sts": 0.7513622883897111,
"eval_sickr_spearman": 0.6895086034818702,
"eval_stsb_spearman": 0.8132159732975519,
"step": 11125
},
{
"epoch": 0.72,
"eval_avg_sts": 0.7524578381239166,
"eval_sickr_spearman": 0.6914134689338043,
"eval_stsb_spearman": 0.813502207314029,
"step": 11250
},
{
"epoch": 0.73,
"eval_avg_sts": 0.7529077801461861,
"eval_sickr_spearman": 0.7008521567396107,
"eval_stsb_spearman": 0.8049634035527615,
"step": 11375
},
{
"electra_acc": 0.9069,
"electra_fix_acc": 0.9903,
"electra_rep_acc": 0.2207,
"epoch": 0.74,
"learning_rate": 6.320235504927685e-06,
"loss": 0.0011,
"neg_sim": -0.0141,
"pos_sim": 0.8003,
"step": 11500
},
{
"epoch": 0.74,
"eval_avg_sts": 0.7520009086329529,
"eval_sickr_spearman": 0.6998127156733884,
"eval_stsb_spearman": 0.8041891015925174,
"step": 11500
},
{
"epoch": 0.74,
"eval_avg_sts": 0.7528385713572732,
"eval_sickr_spearman": 0.6952667640466769,
"eval_stsb_spearman": 0.8104103786678696,
"step": 11625
},
{
"epoch": 0.75,
"eval_avg_sts": 0.7543110510513436,
"eval_sickr_spearman": 0.6966553431892527,
"eval_stsb_spearman": 0.8119667589134344,
"step": 11750
},
{
"epoch": 0.76,
"eval_avg_sts": 0.7483201703337362,
"eval_sickr_spearman": 0.6935844266877055,
"eval_stsb_spearman": 0.8030559139797669,
"step": 11875
},
{
"electra_acc": 0.9071,
"electra_fix_acc": 0.99,
"electra_rep_acc": 0.2271,
"epoch": 0.77,
"learning_rate": 6.1602457442723675e-06,
"loss": 0.0011,
"neg_sim": -0.0141,
"pos_sim": 0.7871,
"step": 12000
},
{
"epoch": 0.77,
"eval_avg_sts": 0.751054540493113,
"eval_sickr_spearman": 0.692834276945445,
"eval_stsb_spearman": 0.809274804040781,
"step": 12000
},
{
"epoch": 0.78,
"eval_avg_sts": 0.7490596138849832,
"eval_sickr_spearman": 0.6888325657291288,
"eval_stsb_spearman": 0.8092866620408378,
"step": 12125
},
{
"epoch": 0.78,
"eval_avg_sts": 0.7507815525034611,
"eval_sickr_spearman": 0.6925058883049127,
"eval_stsb_spearman": 0.8090572167020096,
"step": 12250
},
{
"epoch": 0.79,
"eval_avg_sts": 0.7527821368803079,
"eval_sickr_spearman": 0.6942120971212987,
"eval_stsb_spearman": 0.8113521766393171,
"step": 12375
},
{
"electra_acc": 0.9072,
"electra_fix_acc": 0.99,
"electra_rep_acc": 0.226,
"epoch": 0.8,
"learning_rate": 6.000255983617049e-06,
"loss": 0.0012,
"neg_sim": -0.014,
"pos_sim": 0.7847,
"step": 12500
},
{
"epoch": 0.8,
"eval_avg_sts": 0.7522991659637694,
"eval_sickr_spearman": 0.6944025884696023,
"eval_stsb_spearman": 0.8101957434579367,
"step": 12500
},
{
"epoch": 0.81,
"eval_avg_sts": 0.7521192684266835,
"eval_sickr_spearman": 0.6942240568655569,
"eval_stsb_spearman": 0.8100144799878102,
"step": 12625
},
{
"epoch": 0.82,
"eval_avg_sts": 0.7503450494614133,
"eval_sickr_spearman": 0.6922287968807163,
"eval_stsb_spearman": 0.8084613020421104,
"step": 12750
},
{
"epoch": 0.82,
"eval_avg_sts": 0.7517083341962134,
"eval_sickr_spearman": 0.6941311647153756,
"eval_stsb_spearman": 0.8092855036770511,
"step": 12875
},
{
"electra_acc": 0.9071,
"electra_fix_acc": 0.9901,
"electra_rep_acc": 0.2276,
"epoch": 0.83,
"learning_rate": 5.840266222961732e-06,
"loss": 0.0011,
"neg_sim": -0.0141,
"pos_sim": 0.798,
"step": 13000
},
{
"epoch": 0.83,
"eval_avg_sts": 0.7504582415898822,
"eval_sickr_spearman": 0.6935004202912902,
"eval_stsb_spearman": 0.8074160628884742,
"step": 13000
},
{
"epoch": 0.84,
"eval_avg_sts": 0.7504838292265648,
"eval_sickr_spearman": 0.6931516664637482,
"eval_stsb_spearman": 0.8078159919893814,
"step": 13125
},
{
"epoch": 0.85,
"eval_avg_sts": 0.7511643537875756,
"eval_sickr_spearman": 0.6905446509130283,
"eval_stsb_spearman": 0.8117840566621228,
"step": 13250
},
{
"epoch": 0.86,
"eval_avg_sts": 0.7516853154108011,
"eval_sickr_spearman": 0.6920386417501228,
"eval_stsb_spearman": 0.8113319890714794,
"step": 13375
},
{
"electra_acc": 0.9078,
"electra_fix_acc": 0.9901,
"electra_rep_acc": 0.2307,
"epoch": 0.86,
"learning_rate": 5.680276462306413e-06,
"loss": 0.0011,
"neg_sim": -0.0141,
"pos_sim": 0.7933,
"step": 13500
},
{
"epoch": 0.86,
"eval_avg_sts": 0.7513575139583819,
"eval_sickr_spearman": 0.6917417615121337,
"eval_stsb_spearman": 0.8109732664046301,
"step": 13500
},
{
"epoch": 0.87,
"eval_avg_sts": 0.751417134937681,
"eval_sickr_spearman": 0.6917422418231481,
"eval_stsb_spearman": 0.8110920280522138,
"step": 13625
},
{
"epoch": 0.88,
"eval_avg_sts": 0.7526199054172857,
"eval_sickr_spearman": 0.6928212605169554,
"eval_stsb_spearman": 0.812418550317616,
"step": 13750
},
{
"epoch": 0.89,
"eval_avg_sts": 0.7532239951741806,
"eval_sickr_spearman": 0.693511563506824,
"eval_stsb_spearman": 0.8129364268415373,
"step": 13875
},
{
"electra_acc": 0.9078,
"electra_fix_acc": 0.9899,
"electra_rep_acc": 0.2332,
"epoch": 0.9,
"learning_rate": 5.520286701651095e-06,
"loss": 0.001,
"neg_sim": -0.0142,
"pos_sim": 0.8032,
"step": 14000
},
{
"epoch": 0.9,
"eval_avg_sts": 0.7535878459898898,
"eval_sickr_spearman": 0.6938975894690821,
"eval_stsb_spearman": 0.8132781025106975,
"step": 14000
},
{
"epoch": 0.9,
"eval_avg_sts": 0.7579602803336107,
"eval_sickr_spearman": 0.6938505189896726,
"eval_stsb_spearman": 0.8220700416775488,
"step": 14125
},
{
"epoch": 0.91,
"eval_avg_sts": 0.7582493904128486,
"eval_sickr_spearman": 0.6947031190713006,
"eval_stsb_spearman": 0.8217956617543966,
"step": 14250
},
{
"epoch": 0.92,
"eval_avg_sts": 0.7585045342871344,
"eval_sickr_spearman": 0.6946038387846281,
"eval_stsb_spearman": 0.8224052297896408,
"step": 14375
},
{
"electra_acc": 0.9081,
"electra_fix_acc": 0.99,
"electra_rep_acc": 0.2354,
"epoch": 0.93,
"learning_rate": 5.3602969409957775e-06,
"loss": 0.001,
"neg_sim": -0.0142,
"pos_sim": 0.8056,
"step": 14500
},
{
"epoch": 0.93,
"eval_avg_sts": 0.7586329610292095,
"eval_sickr_spearman": 0.6949082118744413,
"eval_stsb_spearman": 0.8223577101839777,
"step": 14500
},
{
"epoch": 0.94,
"eval_avg_sts": 0.7580983862167923,
"eval_sickr_spearman": 0.6947662799696918,
"eval_stsb_spearman": 0.8214304924638928,
"step": 14625
},
{
"epoch": 0.94,
"eval_avg_sts": 0.7543501652245072,
"eval_sickr_spearman": 0.6945084970482733,
"eval_stsb_spearman": 0.814191833400741,
"step": 14750
},
{
"epoch": 0.95,
"eval_avg_sts": 0.753735401541382,
"eval_sickr_spearman": 0.6931743297643269,
"eval_stsb_spearman": 0.8142964733184371,
"step": 14875
},
{
"electra_acc": 0.9081,
"electra_fix_acc": 0.9899,
"electra_rep_acc": 0.2356,
"epoch": 0.96,
"learning_rate": 5.200307180340458e-06,
"loss": 0.001,
"neg_sim": -0.0142,
"pos_sim": 0.8068,
"step": 15000
},
{
"epoch": 0.96,
"eval_avg_sts": 0.7565776728970167,
"eval_sickr_spearman": 0.699184805084288,
"eval_stsb_spearman": 0.8139705407097454,
"step": 15000
},
{
"epoch": 0.97,
"eval_avg_sts": 0.7555687559765775,
"eval_sickr_spearman": 0.6981326357761847,
"eval_stsb_spearman": 0.8130048761769703,
"step": 15125
},
{
"epoch": 0.98,
"eval_avg_sts": 0.7607557679069236,
"eval_sickr_spearman": 0.7007437985747662,
"eval_stsb_spearman": 0.8207677372390809,
"step": 15250
},
{
"epoch": 0.98,
"eval_avg_sts": 0.7560246287263773,
"eval_sickr_spearman": 0.6948236771359103,
"eval_stsb_spearman": 0.8172255803168443,
"step": 15375
},
{
"electra_acc": 0.9077,
"electra_fix_acc": 0.9897,
"electra_rep_acc": 0.24,
"epoch": 0.99,
"learning_rate": 5.04031741968514e-06,
"loss": 0.0011,
"neg_sim": -0.0141,
"pos_sim": 0.7969,
"step": 15500
},
{
"epoch": 0.99,
"eval_avg_sts": 0.7559485354838383,
"eval_sickr_spearman": 0.6950054268237521,
"eval_stsb_spearman": 0.8168916441439246,
"step": 15500
},
{
"epoch": 1.0,
"eval_avg_sts": 0.7502675969292933,
"eval_sickr_spearman": 0.6866761613989633,
"eval_stsb_spearman": 0.8138590324596235,
"step": 15625
},
{
"epoch": 1.01,
"eval_avg_sts": 0.7497139540559181,
"eval_sickr_spearman": 0.6875880798908671,
"eval_stsb_spearman": 0.811839828220969,
"step": 15750
},
{
"epoch": 1.02,
"eval_avg_sts": 0.7507851484505732,
"eval_sickr_spearman": 0.6884938984328886,
"eval_stsb_spearman": 0.8130763984682579,
"step": 15875
},
{
"electra_acc": 0.9081,
"electra_fix_acc": 0.9898,
"electra_rep_acc": 0.239,
"epoch": 1.02,
"learning_rate": 4.8803276590298225e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.7953,
"step": 16000
},
{
"epoch": 1.02,
"eval_avg_sts": 0.7479678946675965,
"eval_sickr_spearman": 0.6853699555953537,
"eval_stsb_spearman": 0.8105658337398391,
"step": 16000
},
{
"epoch": 1.03,
"eval_avg_sts": 0.7479128193954279,
"eval_sickr_spearman": 0.6867578623025095,
"eval_stsb_spearman": 0.8090677764883464,
"step": 16125
},
{
"epoch": 1.04,
"eval_avg_sts": 0.7487959084542308,
"eval_sickr_spearman": 0.68736439905147,
"eval_stsb_spearman": 0.8102274178569916,
"step": 16250
},
{
"epoch": 1.05,
"eval_avg_sts": 0.7500295795078514,
"eval_sickr_spearman": 0.688809030489424,
"eval_stsb_spearman": 0.8112501285262789,
"step": 16375
},
{
"electra_acc": 0.9078,
"electra_fix_acc": 0.9898,
"electra_rep_acc": 0.2379,
"epoch": 1.06,
"learning_rate": 4.720337898374505e-06,
"loss": 0.0011,
"neg_sim": NaN,
"pos_sim": 0.8055,
"step": 16500
},
{
"epoch": 1.06,
"eval_avg_sts": 0.7515077711907536,
"eval_sickr_spearman": 0.690066388662871,
"eval_stsb_spearman": 0.8129491537186362,
"step": 16500
},
{
"epoch": 1.06,
"eval_avg_sts": 0.7490196746582567,
"eval_sickr_spearman": 0.6842396396852107,
"eval_stsb_spearman": 0.8137997096313027,
"step": 16625
},
{
"epoch": 1.07,
"eval_avg_sts": 0.7500539670025419,
"eval_sickr_spearman": 0.6851718753330229,
"eval_stsb_spearman": 0.8149360586720608,
"step": 16750
},
{
"epoch": 1.08,
"eval_avg_sts": 0.7490200514710061,
"eval_sickr_spearman": 0.6843091406889915,
"eval_stsb_spearman": 0.8137309622530207,
"step": 16875
},
{
"electra_acc": 0.9084,
"electra_fix_acc": 0.9895,
"electra_rep_acc": 0.2437,
"epoch": 1.09,
"learning_rate": 4.560348137719187e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8007,
"step": 17000
},
{
"epoch": 1.09,
"eval_avg_sts": 0.7434034453107932,
"eval_sickr_spearman": 0.677096502310436,
"eval_stsb_spearman": 0.8097103883111502,
"step": 17000
},
{
"epoch": 1.1,
"eval_avg_sts": 0.7431564820160605,
"eval_sickr_spearman": 0.677125561126806,
"eval_stsb_spearman": 0.8091874029053152,
"step": 17125
},
{
"epoch": 1.1,
"eval_avg_sts": 0.7442892283774005,
"eval_sickr_spearman": 0.6786750444591996,
"eval_stsb_spearman": 0.8099034122956015,
"step": 17250
},
{
"epoch": 1.11,
"eval_avg_sts": 0.7452452513650849,
"eval_sickr_spearman": 0.6798861967130629,
"eval_stsb_spearman": 0.8106043060171069,
"step": 17375
},
{
"electra_acc": 0.9085,
"electra_fix_acc": 0.9896,
"electra_rep_acc": 0.2435,
"epoch": 1.12,
"learning_rate": 4.400358377063868e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8058,
"step": 17500
},
{
"epoch": 1.12,
"eval_avg_sts": 0.7457855940878162,
"eval_sickr_spearman": 0.6806044538039685,
"eval_stsb_spearman": 0.810966734371664,
"step": 17500
},
{
"epoch": 1.13,
"eval_avg_sts": 0.7416386948034206,
"eval_sickr_spearman": 0.6797112674416254,
"eval_stsb_spearman": 0.8035661221652158,
"step": 17625
},
{
"epoch": 1.14,
"eval_avg_sts": 0.7423254082827856,
"eval_sickr_spearman": 0.6808730437232104,
"eval_stsb_spearman": 0.8037777728423607,
"step": 17750
},
{
"epoch": 1.14,
"eval_avg_sts": 0.7413901886271141,
"eval_sickr_spearman": 0.681068914554875,
"eval_stsb_spearman": 0.8017114626993534,
"step": 17875
},
{
"electra_acc": 0.9082,
"electra_fix_acc": 0.9892,
"electra_rep_acc": 0.247,
"epoch": 1.15,
"learning_rate": 4.24036861640855e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8062,
"step": 18000
},
{
"epoch": 1.15,
"eval_avg_sts": 0.7429114158358341,
"eval_sickr_spearman": 0.6825547086467617,
"eval_stsb_spearman": 0.8032681230249066,
"step": 18000
},
{
"epoch": 1.16,
"eval_avg_sts": 0.7439526113533993,
"eval_sickr_spearman": 0.6837044771529871,
"eval_stsb_spearman": 0.8042007455538115,
"step": 18125
},
{
"epoch": 1.17,
"eval_avg_sts": 0.7450607408838441,
"eval_sickr_spearman": 0.686960841737187,
"eval_stsb_spearman": 0.8031606400305011,
"step": 18250
},
{
"epoch": 1.18,
"eval_avg_sts": 0.745855277528227,
"eval_sickr_spearman": 0.6876279457050609,
"eval_stsb_spearman": 0.804082609351393,
"step": 18375
},
{
"electra_acc": 0.9078,
"electra_fix_acc": 0.9895,
"electra_rep_acc": 0.242,
"epoch": 1.18,
"learning_rate": 4.080378855753232e-06,
"loss": 0.0011,
"neg_sim": NaN,
"pos_sim": 0.8079,
"step": 18500
},
{
"epoch": 1.18,
"eval_avg_sts": 0.7462797427639389,
"eval_sickr_spearman": 0.6877381290517599,
"eval_stsb_spearman": 0.804821356476118,
"step": 18500
},
{
"epoch": 1.19,
"eval_avg_sts": 0.744680795271585,
"eval_sickr_spearman": 0.6867351916226306,
"eval_stsb_spearman": 0.8026263989205394,
"step": 18625
},
{
"epoch": 1.2,
"eval_avg_sts": 0.7415540412607211,
"eval_sickr_spearman": 0.6830318976395493,
"eval_stsb_spearman": 0.800076184881893,
"step": 18750
},
{
"epoch": 1.21,
"eval_avg_sts": 0.7426288639389378,
"eval_sickr_spearman": 0.6841978045958581,
"eval_stsb_spearman": 0.8010599232820175,
"step": 18875
},
{
"electra_acc": 0.9088,
"electra_fix_acc": 0.9896,
"electra_rep_acc": 0.2456,
"epoch": 1.22,
"learning_rate": 3.920389095097914e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8107,
"step": 19000
},
{
"epoch": 1.22,
"eval_avg_sts": 0.7440228503017443,
"eval_sickr_spearman": 0.6852098768089162,
"eval_stsb_spearman": 0.8028358237945724,
"step": 19000
},
{
"epoch": 1.22,
"eval_avg_sts": 0.7446803355301828,
"eval_sickr_spearman": 0.6853459400446347,
"eval_stsb_spearman": 0.8040147310157307,
"step": 19125
},
{
"epoch": 1.23,
"eval_avg_sts": 0.745178310449073,
"eval_sickr_spearman": 0.685846664277127,
"eval_stsb_spearman": 0.804509956621019,
"step": 19250
},
{
"epoch": 1.24,
"eval_avg_sts": 0.7501895198558063,
"eval_sickr_spearman": 0.6936651189381215,
"eval_stsb_spearman": 0.806713920773491,
"step": 19375
},
{
"electra_acc": 0.9087,
"electra_fix_acc": 0.9894,
"electra_rep_acc": 0.2472,
"epoch": 1.25,
"learning_rate": 3.760399334442596e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.815,
"step": 19500
},
{
"epoch": 1.25,
"eval_avg_sts": 0.7504868457577191,
"eval_sickr_spearman": 0.693660123703572,
"eval_stsb_spearman": 0.8073135678118661,
"step": 19500
},
{
"epoch": 1.26,
"eval_avg_sts": 0.7515054770519676,
"eval_sickr_spearman": 0.6934854345876417,
"eval_stsb_spearman": 0.8095255195162936,
"step": 19625
},
{
"epoch": 1.26,
"eval_avg_sts": 0.7519980000949484,
"eval_sickr_spearman": 0.6937166563099646,
"eval_stsb_spearman": 0.8102793438799323,
"step": 19750
},
{
"epoch": 1.27,
"eval_avg_sts": 0.7521186477822681,
"eval_sickr_spearman": 0.693651237949806,
"eval_stsb_spearman": 0.8105860576147302,
"step": 19875
},
{
"electra_acc": 0.9094,
"electra_fix_acc": 0.9895,
"electra_rep_acc": 0.2499,
"epoch": 1.28,
"learning_rate": 3.600409573787278e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8155,
"step": 20000
},
{
"epoch": 1.28,
"eval_avg_sts": 0.752341790734355,
"eval_sickr_spearman": 0.6935850991231257,
"eval_stsb_spearman": 0.8110984823455842,
"step": 20000
},
{
"epoch": 1.29,
"eval_avg_sts": 0.7524918475690561,
"eval_sickr_spearman": 0.6935691527974482,
"eval_stsb_spearman": 0.811414542340664,
"step": 20125
},
{
"epoch": 1.3,
"eval_avg_sts": 0.7557856359443814,
"eval_sickr_spearman": 0.6970248944837176,
"eval_stsb_spearman": 0.8145463774050452,
"step": 20250
},
{
"epoch": 1.3,
"eval_avg_sts": 0.7552826569601827,
"eval_sickr_spearman": 0.6959100445882376,
"eval_stsb_spearman": 0.814655269332128,
"step": 20375
},
{
"electra_acc": 0.9094,
"electra_fix_acc": 0.9895,
"electra_rep_acc": 0.2515,
"epoch": 1.31,
"learning_rate": 3.44041981313196e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8179,
"step": 20500
},
{
"epoch": 1.31,
"eval_avg_sts": 0.7551388763512048,
"eval_sickr_spearman": 0.6953964480205598,
"eval_stsb_spearman": 0.8148813046818498,
"step": 20500
},
{
"epoch": 1.32,
"eval_avg_sts": 0.7551894637379806,
"eval_sickr_spearman": 0.6956948652537948,
"eval_stsb_spearman": 0.8146840622221664,
"step": 20625
},
{
"epoch": 1.33,
"eval_avg_sts": 0.7549148751641654,
"eval_sickr_spearman": 0.6951038425505989,
"eval_stsb_spearman": 0.8147259077777321,
"step": 20750
},
{
"epoch": 1.34,
"eval_avg_sts": 0.754268177606626,
"eval_sickr_spearman": 0.6945316000080651,
"eval_stsb_spearman": 0.814004755205187,
"step": 20875
},
{
"electra_acc": 0.9089,
"electra_fix_acc": 0.9891,
"electra_rep_acc": 0.2537,
"epoch": 1.34,
"learning_rate": 3.280430052476642e-06,
"loss": 0.0009,
"neg_sim": NaN,
"pos_sim": 0.8193,
"step": 21000
},
{
"epoch": 1.34,
"eval_avg_sts": 0.7544681011807751,
"eval_sickr_spearman": 0.6945047025912598,
"eval_stsb_spearman": 0.8144314997702904,
"step": 21000
},
{
"epoch": 1.35,
"eval_avg_sts": 0.7543169348596496,
"eval_sickr_spearman": 0.694174296644467,
"eval_stsb_spearman": 0.8144595730748322,
"step": 21125
},
{
"epoch": 1.36,
"eval_avg_sts": 0.7553283942077516,
"eval_sickr_spearman": 0.6958436656060499,
"eval_stsb_spearman": 0.8148131228094532,
"step": 21250
},
{
"epoch": 1.37,
"eval_avg_sts": 0.7557282247685372,
"eval_sickr_spearman": 0.6962398741618131,
"eval_stsb_spearman": 0.8152165753752613,
"step": 21375
},
{
"electra_acc": 0.9093,
"electra_fix_acc": 0.9893,
"electra_rep_acc": 0.2553,
"epoch": 1.38,
"learning_rate": 3.1204402918213238e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.822,
"step": 21500
},
{
"epoch": 1.38,
"eval_avg_sts": 0.7533899213084664,
"eval_sickr_spearman": 0.6928663617212057,
"eval_stsb_spearman": 0.813913480895727,
"step": 21500
},
{
"epoch": 1.38,
"eval_avg_sts": 0.7534343863779988,
"eval_sickr_spearman": 0.6929559877564893,
"eval_stsb_spearman": 0.8139127849995083,
"step": 21625
},
{
"epoch": 1.39,
"eval_avg_sts": 0.7539297584804894,
"eval_sickr_spearman": 0.6930869205390096,
"eval_stsb_spearman": 0.8147725964219692,
"step": 21750
},
{
"epoch": 1.4,
"eval_avg_sts": 0.7523440794130264,
"eval_sickr_spearman": 0.691800551580294,
"eval_stsb_spearman": 0.8128876072457588,
"step": 21875
},
{
"electra_acc": 0.9091,
"electra_fix_acc": 0.9891,
"electra_rep_acc": 0.253,
"epoch": 1.41,
"learning_rate": 2.960450531166006e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8192,
"step": 22000
},
{
"epoch": 1.41,
"eval_avg_sts": 0.7535926716898982,
"eval_sickr_spearman": 0.696256204736302,
"eval_stsb_spearman": 0.8109291386434943,
"step": 22000
},
{
"epoch": 1.42,
"eval_avg_sts": 0.7527606084226345,
"eval_sickr_spearman": 0.6954414051315059,
"eval_stsb_spearman": 0.8100798117137631,
"step": 22125
},
{
"epoch": 1.42,
"eval_avg_sts": 0.7533114920096069,
"eval_sickr_spearman": 0.695455142026517,
"eval_stsb_spearman": 0.8111678419926969,
"step": 22250
},
{
"epoch": 1.43,
"eval_avg_sts": 0.7519280470002075,
"eval_sickr_spearman": 0.6946230992563046,
"eval_stsb_spearman": 0.8092329947441105,
"step": 22375
},
{
"electra_acc": 0.9093,
"electra_fix_acc": 0.9893,
"electra_rep_acc": 0.2551,
"epoch": 1.44,
"learning_rate": 2.8004607705106875e-06,
"loss": 0.0011,
"neg_sim": NaN,
"pos_sim": 0.8135,
"step": 22500
},
{
"epoch": 1.44,
"eval_avg_sts": 0.7522224828721787,
"eval_sickr_spearman": 0.694889767931489,
"eval_stsb_spearman": 0.8095551978128684,
"step": 22500
},
{
"epoch": 1.45,
"eval_avg_sts": 0.7509487345671295,
"eval_sickr_spearman": 0.6947804011135146,
"eval_stsb_spearman": 0.8071170680207445,
"step": 22625
},
{
"epoch": 1.46,
"eval_avg_sts": 0.7535117851178551,
"eval_sickr_spearman": 0.6966824807615652,
"eval_stsb_spearman": 0.810341089474145,
"step": 22750
},
{
"epoch": 1.46,
"eval_avg_sts": 0.7554057278055278,
"eval_sickr_spearman": 0.6974278273936819,
"eval_stsb_spearman": 0.8133836282173738,
"step": 22875
},
{
"electra_acc": 0.9094,
"electra_fix_acc": 0.9891,
"electra_rep_acc": 0.2557,
"epoch": 1.47,
"learning_rate": 2.640471009855369e-06,
"loss": 0.0012,
"neg_sim": NaN,
"pos_sim": 0.8108,
"step": 23000
},
{
"epoch": 1.47,
"eval_avg_sts": 0.7573042151426209,
"eval_sickr_spearman": 0.6996995063672988,
"eval_stsb_spearman": 0.814908923917943,
"step": 23000
},
{
"epoch": 1.48,
"eval_avg_sts": 0.7572778466663158,
"eval_sickr_spearman": 0.699593405664222,
"eval_stsb_spearman": 0.8149622876684096,
"step": 23125
},
{
"epoch": 1.49,
"eval_avg_sts": 0.7572787295536805,
"eval_sickr_spearman": 0.6995496493308119,
"eval_stsb_spearman": 0.815007809776549,
"step": 23250
},
{
"epoch": 1.5,
"eval_avg_sts": 0.7578655158883527,
"eval_sickr_spearman": 0.7004442285950968,
"eval_stsb_spearman": 0.8152868031816085,
"step": 23375
},
{
"electra_acc": 0.9093,
"electra_fix_acc": 0.9889,
"electra_rep_acc": 0.2565,
"epoch": 1.5,
"learning_rate": 2.4804812492000513e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8119,
"step": 23500
},
{
"epoch": 1.5,
"eval_avg_sts": 0.7580453272234033,
"eval_sickr_spearman": 0.700563201633359,
"eval_stsb_spearman": 0.8155274528134476,
"step": 23500
},
{
"epoch": 1.51,
"eval_avg_sts": 0.7589489492705888,
"eval_sickr_spearman": 0.7011428889966156,
"eval_stsb_spearman": 0.8167550095445619,
"step": 23625
},
{
"epoch": 1.52,
"eval_avg_sts": 0.759282090184078,
"eval_sickr_spearman": 0.7009553275454997,
"eval_stsb_spearman": 0.8176088528226563,
"step": 23750
},
{
"epoch": 1.53,
"eval_avg_sts": 0.7594355007785425,
"eval_sickr_spearman": 0.7010773265431525,
"eval_stsb_spearman": 0.8177936750139324,
"step": 23875
},
{
"electra_acc": 0.9092,
"electra_fix_acc": 0.989,
"electra_rep_acc": 0.2555,
"epoch": 1.54,
"learning_rate": 2.3204914885447333e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8149,
"step": 24000
},
{
"epoch": 1.54,
"eval_avg_sts": 0.7594249307434521,
"eval_sickr_spearman": 0.7010194490659197,
"eval_stsb_spearman": 0.8178304124209844,
"step": 24000
},
{
"epoch": 1.54,
"eval_avg_sts": 0.7594551033154371,
"eval_sickr_spearman": 0.7012371740487385,
"eval_stsb_spearman": 0.8176730325821358,
"step": 24125
},
{
"epoch": 1.55,
"eval_avg_sts": 0.7589237357383787,
"eval_sickr_spearman": 0.7005038351919814,
"eval_stsb_spearman": 0.8173436362847759,
"step": 24250
},
{
"epoch": 1.56,
"eval_avg_sts": 0.7588819689189907,
"eval_sickr_spearman": 0.7008921666471086,
"eval_stsb_spearman": 0.816871771190873,
"step": 24375
},
{
"electra_acc": 0.9094,
"electra_fix_acc": 0.989,
"electra_rep_acc": 0.2584,
"epoch": 1.57,
"learning_rate": 2.1605017278894154e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8167,
"step": 24500
},
{
"epoch": 1.57,
"eval_avg_sts": 0.7589559409437466,
"eval_sickr_spearman": 0.7010030224292277,
"eval_stsb_spearman": 0.8169088594582655,
"step": 24500
},
{
"epoch": 1.58,
"eval_avg_sts": 0.7589222814938349,
"eval_sickr_spearman": 0.7009486992535012,
"eval_stsb_spearman": 0.8168958637341686,
"step": 24625
},
{
"epoch": 1.58,
"eval_avg_sts": 0.7574752410333385,
"eval_sickr_spearman": 0.6996315903898653,
"eval_stsb_spearman": 0.8153188916768117,
"step": 24750
},
{
"epoch": 1.59,
"eval_avg_sts": 0.7564910076346855,
"eval_sickr_spearman": 0.6992222213123084,
"eval_stsb_spearman": 0.8137597939570625,
"step": 24875
},
{
"electra_acc": 0.9093,
"electra_fix_acc": 0.9891,
"electra_rep_acc": 0.2566,
"epoch": 1.6,
"learning_rate": 2.000511967234097e-06,
"loss": 0.0011,
"neg_sim": NaN,
"pos_sim": 0.8178,
"step": 25000
},
{
"epoch": 1.6,
"eval_avg_sts": 0.754858122417589,
"eval_sickr_spearman": 0.6972444446483912,
"eval_stsb_spearman": 0.8124718001867869,
"step": 25000
},
{
"epoch": 1.61,
"eval_avg_sts": 0.7550451133191622,
"eval_sickr_spearman": 0.6974792206722207,
"eval_stsb_spearman": 0.8126110059661038,
"step": 25125
},
{
"epoch": 1.62,
"eval_avg_sts": 0.754828720368155,
"eval_sickr_spearman": 0.6962917477513663,
"eval_stsb_spearman": 0.8133656929849435,
"step": 25250
},
{
"epoch": 1.62,
"eval_avg_sts": 0.7548476950408536,
"eval_sickr_spearman": 0.6961924674646935,
"eval_stsb_spearman": 0.8135029226170137,
"step": 25375
},
{
"electra_acc": 0.9096,
"electra_fix_acc": 0.9892,
"electra_rep_acc": 0.2585,
"epoch": 1.63,
"learning_rate": 1.8405222065787792e-06,
"loss": 0.0011,
"neg_sim": NaN,
"pos_sim": 0.8169,
"step": 25500
},
{
"epoch": 1.63,
"eval_avg_sts": 0.7530610693683093,
"eval_sickr_spearman": 0.6936729960387574,
"eval_stsb_spearman": 0.8124491426978612,
"step": 25500
},
{
"epoch": 1.64,
"eval_avg_sts": 0.7537427601811764,
"eval_sickr_spearman": 0.6928838450421292,
"eval_stsb_spearman": 0.8146016753202237,
"step": 25625
},
{
"epoch": 1.65,
"eval_avg_sts": 0.7538542984125725,
"eval_sickr_spearman": 0.6928155928469857,
"eval_stsb_spearman": 0.8148930039781592,
"step": 25750
},
{
"epoch": 1.66,
"eval_avg_sts": 0.7539219581746606,
"eval_sickr_spearman": 0.6929616554264589,
"eval_stsb_spearman": 0.8148822609228621,
"step": 25875
},
{
"electra_acc": 0.9096,
"electra_fix_acc": 0.9892,
"electra_rep_acc": 0.2568,
"epoch": 1.66,
"learning_rate": 1.6805324459234608e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8191,
"step": 26000
},
{
"epoch": 1.66,
"eval_avg_sts": 0.7537628256146356,
"eval_sickr_spearman": 0.6923114103751897,
"eval_stsb_spearman": 0.8152142408540815,
"step": 26000
},
{
"epoch": 1.67,
"eval_avg_sts": 0.754620615839658,
"eval_sickr_spearman": 0.6932064219193875,
"eval_stsb_spearman": 0.8160348097599284,
"step": 26125
},
{
"epoch": 1.68,
"eval_avg_sts": 0.7518830889314513,
"eval_sickr_spearman": 0.690710149515446,
"eval_stsb_spearman": 0.8130560283474566,
"step": 26250
},
{
"epoch": 1.69,
"eval_avg_sts": 0.7509225404957667,
"eval_sickr_spearman": 0.6898265693733905,
"eval_stsb_spearman": 0.8120185116181428,
"step": 26375
},
{
"electra_acc": 0.9096,
"electra_fix_acc": 0.9892,
"electra_rep_acc": 0.256,
"epoch": 1.7,
"learning_rate": 1.520542685268143e-06,
"loss": 0.0011,
"neg_sim": NaN,
"pos_sim": 0.8194,
"step": 26500
},
{
"epoch": 1.7,
"eval_avg_sts": 0.7509117147704546,
"eval_sickr_spearman": 0.690458994886026,
"eval_stsb_spearman": 0.8113644346548832,
"step": 26500
},
{
"epoch": 1.7,
"eval_avg_sts": 0.751325870403242,
"eval_sickr_spearman": 0.6910303728687339,
"eval_stsb_spearman": 0.8116213679377499,
"step": 26625
},
{
"epoch": 1.71,
"eval_avg_sts": 0.7505685715846826,
"eval_sickr_spearman": 0.6889766590334431,
"eval_stsb_spearman": 0.812160484135922,
"step": 26750
},
{
"epoch": 1.72,
"eval_avg_sts": 0.7505717215143468,
"eval_sickr_spearman": 0.6889912124571789,
"eval_stsb_spearman": 0.8121522305715146,
"step": 26875
},
{
"electra_acc": 0.9097,
"electra_fix_acc": 0.989,
"electra_rep_acc": 0.2597,
"epoch": 1.73,
"learning_rate": 1.3605529246128248e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8193,
"step": 27000
},
{
"epoch": 1.73,
"eval_avg_sts": 0.7506556450635647,
"eval_sickr_spearman": 0.6890205594601575,
"eval_stsb_spearman": 0.812290730666972,
"step": 27000
},
{
"epoch": 1.74,
"eval_avg_sts": 0.7509018804381178,
"eval_sickr_spearman": 0.6900798373712735,
"eval_stsb_spearman": 0.811723923504962,
"step": 27125
},
{
"epoch": 1.74,
"eval_avg_sts": 0.7500895498195226,
"eval_sickr_spearman": 0.6892968343556296,
"eval_stsb_spearman": 0.8108822652834156,
"step": 27250
},
{
"epoch": 1.75,
"eval_avg_sts": 0.7506787202343813,
"eval_sickr_spearman": 0.6893182081957694,
"eval_stsb_spearman": 0.8120392322729932,
"step": 27375
},
{
"electra_acc": 0.9101,
"electra_fix_acc": 0.989,
"electra_rep_acc": 0.262,
"epoch": 1.76,
"learning_rate": 1.2005631639575069e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8192,
"step": 27500
},
{
"epoch": 1.76,
"eval_avg_sts": 0.7507236365099843,
"eval_sickr_spearman": 0.689373636086829,
"eval_stsb_spearman": 0.8120736369331395,
"step": 27500
},
{
"epoch": 1.77,
"eval_avg_sts": 0.7507778777687245,
"eval_sickr_spearman": 0.6893403025024311,
"eval_stsb_spearman": 0.812215453035018,
"step": 27625
},
{
"epoch": 1.78,
"eval_avg_sts": 0.7508593494865333,
"eval_sickr_spearman": 0.6894537519640278,
"eval_stsb_spearman": 0.8122649470090388,
"step": 27750
},
{
"epoch": 1.78,
"eval_avg_sts": 0.7509704306691856,
"eval_sickr_spearman": 0.6894628778733011,
"eval_stsb_spearman": 0.8124779834650699,
"step": 27875
},
{
"electra_acc": 0.9101,
"electra_fix_acc": 0.9891,
"electra_rep_acc": 0.2586,
"epoch": 1.79,
"learning_rate": 1.0405734033021888e-06,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8201,
"step": 28000
},
{
"epoch": 1.79,
"eval_avg_sts": 0.7509106758642159,
"eval_sickr_spearman": 0.6893112436860609,
"eval_stsb_spearman": 0.812510108042371,
"step": 28000
},
{
"epoch": 1.8,
"eval_avg_sts": 0.7508459705581159,
"eval_sickr_spearman": 0.6899334385740902,
"eval_stsb_spearman": 0.8117585025421415,
"step": 28125
},
{
"epoch": 1.81,
"eval_avg_sts": 0.7510477486103114,
"eval_sickr_spearman": 0.6902538540517839,
"eval_stsb_spearman": 0.811841643168839,
"step": 28250
},
{
"epoch": 1.82,
"eval_avg_sts": 0.7507757787312187,
"eval_sickr_spearman": 0.6898934766976937,
"eval_stsb_spearman": 0.8116580807647437,
"step": 28375
},
{
"electra_acc": 0.9098,
"electra_fix_acc": 0.989,
"electra_rep_acc": 0.2594,
"epoch": 1.82,
"learning_rate": 8.805836426468707e-07,
"loss": 0.0011,
"neg_sim": NaN,
"pos_sim": 0.8208,
"step": 28500
},
{
"epoch": 1.82,
"eval_avg_sts": 0.7507727971903551,
"eval_sickr_spearman": 0.6899099993965885,
"eval_stsb_spearman": 0.8116355949841215,
"step": 28500
},
{
"epoch": 1.83,
"eval_avg_sts": 0.7508146549871678,
"eval_sickr_spearman": 0.6899898270871786,
"eval_stsb_spearman": 0.811639482887157,
"step": 28625
},
{
"epoch": 1.84,
"eval_avg_sts": 0.7506882515044668,
"eval_sickr_spearman": 0.6898797398026825,
"eval_stsb_spearman": 0.8114967632062512,
"step": 28750
},
{
"epoch": 1.85,
"eval_avg_sts": 0.7507604828702781,
"eval_sickr_spearman": 0.6899782035606307,
"eval_stsb_spearman": 0.8115427621799256,
"step": 28875
},
{
"electra_acc": 0.9101,
"electra_fix_acc": 0.989,
"electra_rep_acc": 0.2588,
"epoch": 1.86,
"learning_rate": 7.205938819915525e-07,
"loss": 0.0009,
"neg_sim": NaN,
"pos_sim": 0.821,
"step": 29000
},
{
"epoch": 1.86,
"eval_avg_sts": 0.7507686428892963,
"eval_sickr_spearman": 0.6899909798336131,
"eval_stsb_spearman": 0.8115463059449796,
"step": 29000
},
{
"epoch": 1.86,
"eval_avg_sts": 0.7509200158436065,
"eval_sickr_spearman": 0.6902724420880404,
"eval_stsb_spearman": 0.8115675895991724,
"step": 29125
},
{
"epoch": 1.87,
"eval_avg_sts": 0.7507926487881307,
"eval_sickr_spearman": 0.690199482844956,
"eval_stsb_spearman": 0.8113858147313052,
"step": 29250
},
{
"epoch": 1.88,
"eval_avg_sts": 0.7510272796849722,
"eval_sickr_spearman": 0.6903056796102356,
"eval_stsb_spearman": 0.8117488797597087,
"step": 29375
},
{
"electra_acc": 0.91,
"electra_fix_acc": 0.9889,
"electra_rep_acc": 0.2628,
"epoch": 1.89,
"learning_rate": 5.606041213362345e-07,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8219,
"step": 29500
},
{
"epoch": 1.89,
"eval_avg_sts": 0.7516619765140928,
"eval_sickr_spearman": 0.6907379114920772,
"eval_stsb_spearman": 0.8125860415361085,
"step": 29500
},
{
"epoch": 1.9,
"eval_avg_sts": 0.7516247721458812,
"eval_sickr_spearman": 0.6906019834750073,
"eval_stsb_spearman": 0.812647560816755,
"step": 29625
},
{
"epoch": 1.9,
"eval_avg_sts": 0.7515302759169064,
"eval_sickr_spearman": 0.6904020780308219,
"eval_stsb_spearman": 0.8126584738029908,
"step": 29750
},
{
"epoch": 1.91,
"eval_avg_sts": 0.7512737585285767,
"eval_sickr_spearman": 0.6900924695509518,
"eval_stsb_spearman": 0.8124550475062015,
"step": 29875
},
{
"electra_acc": 0.9095,
"electra_fix_acc": 0.9889,
"electra_rep_acc": 0.2603,
"epoch": 1.92,
"learning_rate": 4.0061436068091647e-07,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8219,
"step": 30000
},
{
"epoch": 1.92,
"eval_avg_sts": 0.751445438691633,
"eval_sickr_spearman": 0.6902850742677187,
"eval_stsb_spearman": 0.8126058031155474,
"step": 30000
},
{
"epoch": 1.93,
"eval_avg_sts": 0.7514690784500874,
"eval_sickr_spearman": 0.6903169188879721,
"eval_stsb_spearman": 0.8126212380122027,
"step": 30125
},
{
"epoch": 1.94,
"eval_avg_sts": 0.7513646557354139,
"eval_sickr_spearman": 0.6901293094057549,
"eval_stsb_spearman": 0.812600002065073,
"step": 30250
},
{
"epoch": 1.94,
"eval_avg_sts": 0.7513546637483999,
"eval_sickr_spearman": 0.6900938624528935,
"eval_stsb_spearman": 0.8126154650439061,
"step": 30375
},
{
"electra_acc": 0.9098,
"electra_fix_acc": 0.989,
"electra_rep_acc": 0.2615,
"epoch": 1.95,
"learning_rate": 2.4062460002559835e-07,
"loss": 0.0009,
"neg_sim": NaN,
"pos_sim": 0.8227,
"step": 30500
},
{
"epoch": 1.95,
"eval_avg_sts": 0.7513856383156294,
"eval_sickr_spearman": 0.6901277243794074,
"eval_stsb_spearman": 0.8126435522518516,
"step": 30500
},
{
"epoch": 1.96,
"eval_avg_sts": 0.7514646471464337,
"eval_sickr_spearman": 0.6902287818168333,
"eval_stsb_spearman": 0.8127005124760341,
"step": 30625
},
{
"epoch": 1.97,
"eval_avg_sts": 0.7515241048574239,
"eval_sickr_spearman": 0.6903239314287821,
"eval_stsb_spearman": 0.8127242782860657,
"step": 30750
},
{
"epoch": 1.98,
"eval_avg_sts": 0.7517451176255749,
"eval_sickr_spearman": 0.6905863253359386,
"eval_stsb_spearman": 0.8129039099152112,
"step": 30875
},
{
"electra_acc": 0.9096,
"electra_fix_acc": 0.989,
"electra_rep_acc": 0.2613,
"epoch": 1.98,
"learning_rate": 8.06348393702803e-08,
"loss": 0.001,
"neg_sim": NaN,
"pos_sim": 0.8228,
"step": 31000
},
{
"epoch": 1.98,
"eval_avg_sts": 0.7517585752042792,
"eval_sickr_spearman": 0.6906067865851512,
"eval_stsb_spearman": 0.8129103638234072,
"step": 31000
},
{
"epoch": 1.99,
"eval_avg_sts": 0.7517603113599198,
"eval_sickr_spearman": 0.6906109172598749,
"eval_stsb_spearman": 0.8129097054599647,
"step": 31125
},
{
"epoch": 2.0,
"eval_avg_sts": 0.7517610887734341,
"eval_sickr_spearman": 0.6906109172598749,
"eval_stsb_spearman": 0.8129112602869931,
"step": 31250
},
{
"epoch": 2.0,
"step": 31252,
"train_runtime": 14642.7016,
"train_samples_per_second": 2.134
}
],
"max_steps": 31252,
"num_train_epochs": 2,
"total_flos": 571223595241308000,
"trial_name": null,
"trial_params": null
}