SynCSE-partial-RoBERTa-base / trainer_state.json
leoozy's picture
Upload 12 files
a84fbbd
{
"best_metric": 0.8509230445807242,
"best_model_checkpoint": "result/my-sup-simcse-roberta-base_filtered_final_nli_pos_neg",
"epoch": 3.0,
"global_step": 1542,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.05,
"eval_avg_sts": 0.7970172556800076,
"eval_sickr_spearman": 0.75025416187497,
"eval_stsb_spearman": 0.8437803494850452,
"step": 25
},
{
"epoch": 0.1,
"eval_avg_sts": 0.8184611939888191,
"eval_sickr_spearman": 0.787718901307712,
"eval_stsb_spearman": 0.8492034866699262,
"step": 50
},
{
"epoch": 0.15,
"eval_avg_sts": 0.82628745175756,
"eval_sickr_spearman": 0.7880150610791795,
"eval_stsb_spearman": 0.8645598424359405,
"step": 75
},
{
"epoch": 0.19,
"eval_avg_sts": 0.8214663219382123,
"eval_sickr_spearman": 0.7863345008709616,
"eval_stsb_spearman": 0.856598143005463,
"step": 100
},
{
"epoch": 0.24,
"eval_avg_sts": 0.8245359107619952,
"eval_sickr_spearman": 0.7852220525305533,
"eval_stsb_spearman": 0.8638497689934371,
"step": 125
},
{
"epoch": 0.29,
"eval_avg_sts": 0.8369552464117445,
"eval_sickr_spearman": 0.8047760902992264,
"eval_stsb_spearman": 0.8691344025242626,
"step": 150
},
{
"epoch": 0.34,
"eval_avg_sts": 0.8347841673419429,
"eval_sickr_spearman": 0.8065769684165468,
"eval_stsb_spearman": 0.862991366267339,
"step": 175
},
{
"epoch": 0.39,
"eval_avg_sts": 0.8365894758739485,
"eval_sickr_spearman": 0.807168999772873,
"eval_stsb_spearman": 0.866009951975024,
"step": 200
},
{
"epoch": 0.44,
"eval_avg_sts": 0.8373694984089681,
"eval_sickr_spearman": 0.8121874333444314,
"eval_stsb_spearman": 0.8625515634735048,
"step": 225
},
{
"epoch": 0.49,
"eval_avg_sts": 0.8353378734135991,
"eval_sickr_spearman": 0.8028517722512096,
"eval_stsb_spearman": 0.8678239745759886,
"step": 250
},
{
"epoch": 0.54,
"eval_avg_sts": 0.8421542038447196,
"eval_sickr_spearman": 0.8175947747443341,
"eval_stsb_spearman": 0.8667136329451051,
"step": 275
},
{
"epoch": 0.58,
"eval_avg_sts": 0.8360416637090009,
"eval_sickr_spearman": 0.8098268968331538,
"eval_stsb_spearman": 0.8622564305848479,
"step": 300
},
{
"epoch": 0.63,
"eval_avg_sts": 0.8432321348956329,
"eval_sickr_spearman": 0.8172194597176966,
"eval_stsb_spearman": 0.8692448100735692,
"step": 325
},
{
"epoch": 0.68,
"eval_avg_sts": 0.8406118882697353,
"eval_sickr_spearman": 0.8133721684925038,
"eval_stsb_spearman": 0.8678516080469669,
"step": 350
},
{
"epoch": 0.73,
"eval_avg_sts": 0.8364168982681424,
"eval_sickr_spearman": 0.8088663228354931,
"eval_stsb_spearman": 0.8639674737007915,
"step": 375
},
{
"epoch": 0.78,
"eval_avg_sts": 0.8411726277287294,
"eval_sickr_spearman": 0.8214387037924262,
"eval_stsb_spearman": 0.8609065516650325,
"step": 400
},
{
"epoch": 0.83,
"eval_avg_sts": 0.8446725188995174,
"eval_sickr_spearman": 0.824545739682255,
"eval_stsb_spearman": 0.86479929811678,
"step": 425
},
{
"epoch": 0.88,
"eval_avg_sts": 0.846832980887532,
"eval_sickr_spearman": 0.8305859869058076,
"eval_stsb_spearman": 0.8630799748692564,
"step": 450
},
{
"epoch": 0.92,
"eval_avg_sts": 0.8417173844571794,
"eval_sickr_spearman": 0.8217314533556915,
"eval_stsb_spearman": 0.8617033155586672,
"step": 475
},
{
"epoch": 0.97,
"learning_rate": 3.378728923476006e-05,
"loss": 0.5309,
"step": 500
},
{
"epoch": 0.97,
"eval_avg_sts": 0.8394389881737694,
"eval_sickr_spearman": 0.8176411247572218,
"eval_stsb_spearman": 0.8612368515903169,
"step": 500
},
{
"epoch": 1.02,
"eval_avg_sts": 0.8424515638291671,
"eval_sickr_spearman": 0.8244119250336484,
"eval_stsb_spearman": 0.8604912026246858,
"step": 525
},
{
"epoch": 1.07,
"eval_avg_sts": 0.843505143147442,
"eval_sickr_spearman": 0.8292789645734737,
"eval_stsb_spearman": 0.8577313217214102,
"step": 550
},
{
"epoch": 1.12,
"eval_avg_sts": 0.843100574356599,
"eval_sickr_spearman": 0.8253128924344243,
"eval_stsb_spearman": 0.8608882562787737,
"step": 575
},
{
"epoch": 1.17,
"eval_avg_sts": 0.8431532651801806,
"eval_sickr_spearman": 0.8231387646278283,
"eval_stsb_spearman": 0.863167765732533,
"step": 600
},
{
"epoch": 1.22,
"eval_avg_sts": 0.8411549297525207,
"eval_sickr_spearman": 0.8223049447068626,
"eval_stsb_spearman": 0.8600049147981788,
"step": 625
},
{
"epoch": 1.26,
"eval_avg_sts": 0.8444909588234402,
"eval_sickr_spearman": 0.8264434965311761,
"eval_stsb_spearman": 0.8625384211157042,
"step": 650
},
{
"epoch": 1.31,
"eval_avg_sts": 0.8442403156760934,
"eval_sickr_spearman": 0.8282549414908131,
"eval_stsb_spearman": 0.8602256898613736,
"step": 675
},
{
"epoch": 1.36,
"eval_avg_sts": 0.8433546951723836,
"eval_sickr_spearman": 0.8271446545499697,
"eval_stsb_spearman": 0.8595647357947974,
"step": 700
},
{
"epoch": 1.41,
"eval_avg_sts": 0.8420306309738643,
"eval_sickr_spearman": 0.8233070175761659,
"eval_stsb_spearman": 0.8607542443715627,
"step": 725
},
{
"epoch": 1.46,
"eval_avg_sts": 0.8391026032092552,
"eval_sickr_spearman": 0.8208057019065734,
"eval_stsb_spearman": 0.8573995045119369,
"step": 750
},
{
"epoch": 1.51,
"eval_avg_sts": 0.8413993052906461,
"eval_sickr_spearman": 0.8210936963907962,
"eval_stsb_spearman": 0.8617049141904959,
"step": 775
},
{
"epoch": 1.56,
"eval_avg_sts": 0.8434142809317196,
"eval_sickr_spearman": 0.8264181361096167,
"eval_stsb_spearman": 0.8604104257538223,
"step": 800
},
{
"epoch": 1.61,
"eval_avg_sts": 0.8445746067631623,
"eval_sickr_spearman": 0.8273452804606768,
"eval_stsb_spearman": 0.8618039330656478,
"step": 825
},
{
"epoch": 1.65,
"eval_avg_sts": 0.8429633955631959,
"eval_sickr_spearman": 0.8270336546745461,
"eval_stsb_spearman": 0.8588931364518456,
"step": 850
},
{
"epoch": 1.7,
"eval_avg_sts": 0.8472516730528594,
"eval_sickr_spearman": 0.8334014259788056,
"eval_stsb_spearman": 0.8611019201269132,
"step": 875
},
{
"epoch": 1.75,
"eval_avg_sts": 0.8509230445807242,
"eval_sickr_spearman": 0.8376631295472063,
"eval_stsb_spearman": 0.8641829596142421,
"step": 900
},
{
"epoch": 1.8,
"eval_avg_sts": 0.8453984948023106,
"eval_sickr_spearman": 0.82835503830621,
"eval_stsb_spearman": 0.8624419512984111,
"step": 925
},
{
"epoch": 1.85,
"eval_avg_sts": 0.8441717460262391,
"eval_sickr_spearman": 0.8250034760789602,
"eval_stsb_spearman": 0.8633400159735181,
"step": 950
},
{
"epoch": 1.9,
"eval_avg_sts": 0.8480020725945073,
"eval_sickr_spearman": 0.8306723468261933,
"eval_stsb_spearman": 0.8653317983628214,
"step": 975
},
{
"epoch": 1.95,
"learning_rate": 1.7574578469520103e-05,
"loss": 0.2714,
"step": 1000
},
{
"epoch": 1.95,
"eval_avg_sts": 0.8467505999509635,
"eval_sickr_spearman": 0.8330692909123609,
"eval_stsb_spearman": 0.8604319089895659,
"step": 1000
},
{
"epoch": 1.99,
"eval_avg_sts": 0.8494326339830918,
"eval_sickr_spearman": 0.8343606551056261,
"eval_stsb_spearman": 0.8645046128605576,
"step": 1025
},
{
"epoch": 2.04,
"eval_avg_sts": 0.8452826795684555,
"eval_sickr_spearman": 0.8291940455861312,
"eval_stsb_spearman": 0.8613713135507798,
"step": 1050
},
{
"epoch": 2.09,
"eval_avg_sts": 0.8457791155470658,
"eval_sickr_spearman": 0.8285575854609747,
"eval_stsb_spearman": 0.8630006456331568,
"step": 1075
},
{
"epoch": 2.14,
"eval_avg_sts": 0.8447554891917526,
"eval_sickr_spearman": 0.8279863996026727,
"eval_stsb_spearman": 0.8615245787808324,
"step": 1100
},
{
"epoch": 2.19,
"eval_avg_sts": 0.847764082068214,
"eval_sickr_spearman": 0.8317096745239522,
"eval_stsb_spearman": 0.8638184896124759,
"step": 1125
},
{
"epoch": 2.24,
"eval_avg_sts": 0.8451478500577279,
"eval_sickr_spearman": 0.828890681149448,
"eval_stsb_spearman": 0.8614050189660079,
"step": 1150
},
{
"epoch": 2.29,
"eval_avg_sts": 0.8439050801814342,
"eval_sickr_spearman": 0.8240606735888313,
"eval_stsb_spearman": 0.8637494867740371,
"step": 1175
},
{
"epoch": 2.33,
"eval_avg_sts": 0.8447525482867402,
"eval_sickr_spearman": 0.8263380682635195,
"eval_stsb_spearman": 0.863167028309961,
"step": 1200
},
{
"epoch": 2.38,
"eval_avg_sts": 0.8455355460641325,
"eval_sickr_spearman": 0.8271100241258328,
"eval_stsb_spearman": 0.8639610680024322,
"step": 1225
},
{
"epoch": 2.43,
"eval_avg_sts": 0.8467670558910284,
"eval_sickr_spearman": 0.8293558143357747,
"eval_stsb_spearman": 0.8641782974462822,
"step": 1250
},
{
"epoch": 2.48,
"eval_avg_sts": 0.8438803647863737,
"eval_sickr_spearman": 0.8270033470495388,
"eval_stsb_spearman": 0.8607573825232087,
"step": 1275
},
{
"epoch": 2.53,
"eval_avg_sts": 0.8454224221400783,
"eval_sickr_spearman": 0.8282425974977434,
"eval_stsb_spearman": 0.8626022467824133,
"step": 1300
},
{
"epoch": 2.58,
"eval_avg_sts": 0.8449894067547106,
"eval_sickr_spearman": 0.8266484932721141,
"eval_stsb_spearman": 0.8633303202373069,
"step": 1325
},
{
"epoch": 2.63,
"eval_avg_sts": 0.8460181390370703,
"eval_sickr_spearman": 0.8278362063484757,
"eval_stsb_spearman": 0.864200071725665,
"step": 1350
},
{
"epoch": 2.68,
"eval_avg_sts": 0.8465150060489774,
"eval_sickr_spearman": 0.8285329455059368,
"eval_stsb_spearman": 0.8644970665920182,
"step": 1375
},
{
"epoch": 2.72,
"eval_avg_sts": 0.8471917941480929,
"eval_sickr_spearman": 0.8293239697155211,
"eval_stsb_spearman": 0.8650596185806647,
"step": 1400
},
{
"epoch": 2.77,
"eval_avg_sts": 0.8469228159218247,
"eval_sickr_spearman": 0.8292969282054116,
"eval_stsb_spearman": 0.8645487036382378,
"step": 1425
},
{
"epoch": 2.82,
"eval_avg_sts": 0.846386304453965,
"eval_sickr_spearman": 0.8283193992289432,
"eval_stsb_spearman": 0.864453209678987,
"step": 1450
},
{
"epoch": 2.87,
"eval_avg_sts": 0.8456684357631843,
"eval_sickr_spearman": 0.8282710799408963,
"eval_stsb_spearman": 0.8630657915854721,
"step": 1475
},
{
"epoch": 2.92,
"learning_rate": 1.3618677042801557e-06,
"loss": 0.2376,
"step": 1500
},
{
"epoch": 2.92,
"eval_avg_sts": 0.846184384136446,
"eval_sickr_spearman": 0.8285680082099867,
"eval_stsb_spearman": 0.8638007600629054,
"step": 1500
},
{
"epoch": 2.97,
"eval_avg_sts": 0.8465883297884287,
"eval_sickr_spearman": 0.8287545610079723,
"eval_stsb_spearman": 0.864422098568885,
"step": 1525
},
{
"epoch": 3.0,
"step": 1542,
"train_runtime": 2217.7106,
"train_samples_per_second": 0.695
}
],
"max_steps": 1542,
"num_train_epochs": 3,
"total_flos": 148906707659089920,
"trial_name": null,
"trial_params": null
}