longformer-spans / checkpoint-2187 /trainer_state.json
Theoreticallyhugo's picture
Training in progress, epoch 27, checkpoint
84a500f verified
raw
history blame
31.4 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 27.0,
"eval_steps": 500,
"global_step": 2187,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_B": {
"f1-score": 0.8331822302810515,
"precision": 0.7901977644024075,
"recall": 0.8811121764141898,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9471294753175105,
"precision": 0.9443616777446711,
"recall": 0.9499135446685879,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.8992868897421833,
"precision": 0.9103731674811195,
"recall": 0.8884673748103187,
"support": 9226.0
},
"eval_accuracy": 0.9267895289474637,
"eval_loss": 0.20478281378746033,
"eval_macro avg": {
"f1-score": 0.8931995317802484,
"precision": 0.8816442032093993,
"recall": 0.9064976986310321,
"support": 27619.0
},
"eval_runtime": 4.6128,
"eval_samples_per_second": 17.343,
"eval_steps_per_second": 2.168,
"eval_weighted avg": {
"f1-score": 0.9268447919078652,
"precision": 0.9271861479533134,
"recall": 0.9267895289474637,
"support": 27619.0
},
"step": 81
},
{
"epoch": 2.0,
"eval_B": {
"f1-score": 0.8682547399124939,
"precision": 0.8806706114398422,
"recall": 0.8561840843720039,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9590983771942466,
"precision": 0.9592919333448654,
"recall": 0.9589048991354466,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9226525313717006,
"precision": 0.9208594256100194,
"recall": 0.9244526338608281,
"support": 9226.0
},
"eval_accuracy": 0.9435171439950758,
"eval_loss": 0.16574698686599731,
"eval_macro avg": {
"f1-score": 0.9166685494928136,
"precision": 0.9202739901315756,
"recall": 0.9131805391227594,
"support": 27619.0
},
"eval_runtime": 4.6238,
"eval_samples_per_second": 17.302,
"eval_steps_per_second": 2.163,
"eval_weighted avg": {
"f1-score": 0.9434932036816766,
"precision": 0.9434846863370583,
"recall": 0.9435171439950758,
"support": 27619.0
},
"step": 162
},
{
"epoch": 3.0,
"eval_B": {
"f1-score": 0.8791725434884815,
"precision": 0.8625461254612546,
"recall": 0.8964525407478428,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9588372159825352,
"precision": 0.9557349825345015,
"recall": 0.9619596541786744,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9204284621270084,
"precision": 0.9282407407407407,
"recall": 0.9127465857359636,
"support": 9226.0
},
"eval_accuracy": 0.9430464535283681,
"eval_loss": 0.15394224226474762,
"eval_macro avg": {
"f1-score": 0.9194794071993417,
"precision": 0.9155072829121655,
"recall": 0.9237195935541602,
"support": 27619.0
},
"eval_runtime": 4.6233,
"eval_samples_per_second": 17.303,
"eval_steps_per_second": 2.163,
"eval_weighted avg": {
"f1-score": 0.9429985029052194,
"precision": 0.9430314866542512,
"recall": 0.9430464535283681,
"support": 27619.0
},
"step": 243
},
{
"epoch": 4.0,
"eval_B": {
"f1-score": 0.8800746616892207,
"precision": 0.8572727272727273,
"recall": 0.9041227229146692,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9537844036697248,
"precision": 0.9488876212207644,
"recall": 0.9587319884726225,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9100192149327477,
"precision": 0.9220157970853265,
"recall": 0.8983308042488619,
"support": 9226.0
},
"eval_accuracy": 0.9364929939534379,
"eval_loss": 0.19801080226898193,
"eval_macro avg": {
"f1-score": 0.9146260934305644,
"precision": 0.9093920485262728,
"recall": 0.9203951718787179,
"support": 27619.0
},
"eval_runtime": 4.6132,
"eval_samples_per_second": 17.342,
"eval_steps_per_second": 2.168,
"eval_weighted avg": {
"f1-score": 0.9363812792925563,
"precision": 0.9364514800186444,
"recall": 0.9364929939534379,
"support": 27619.0
},
"step": 324
},
{
"epoch": 5.0,
"eval_B": {
"f1-score": 0.883877159309021,
"precision": 0.8847262247838616,
"recall": 0.8830297219558965,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9591500482927107,
"precision": 0.9456643513331839,
"recall": 0.9730259365994236,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9191176470588236,
"precision": 0.9454503781801513,
"recall": 0.8942120095382614,
"support": 9226.0
},
"eval_accuracy": 0.9432999022412107,
"eval_loss": 0.20074255764484406,
"eval_macro avg": {
"f1-score": 0.9207149515535183,
"precision": 0.9252803180990656,
"recall": 0.9167558893645271,
"support": 27619.0
},
"eval_runtime": 4.6092,
"eval_samples_per_second": 17.356,
"eval_steps_per_second": 2.17,
"eval_weighted avg": {
"f1-score": 0.9429348139614955,
"precision": 0.9432916158141273,
"recall": 0.9432999022412107,
"support": 27619.0
},
"step": 405
},
{
"epoch": 6.0,
"eval_B": {
"f1-score": 0.8897338403041826,
"precision": 0.882186616399623,
"recall": 0.8974113135186961,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9624160683661308,
"precision": 0.9707969272268809,
"recall": 0.954178674351585,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9297955261331482,
"precision": 0.9161493950552341,
"recall": 0.9438543247344461,
"support": 9226.0
},
"eval_accuracy": 0.9485861182519281,
"eval_loss": 0.16565027832984924,
"eval_macro avg": {
"f1-score": 0.9273151449344872,
"precision": 0.9230443128939126,
"recall": 0.9318147708682424,
"support": 27619.0
},
"eval_runtime": 4.6219,
"eval_samples_per_second": 17.309,
"eval_steps_per_second": 2.164,
"eval_weighted avg": {
"f1-score": 0.9487745648174827,
"precision": 0.9491959030765336,
"recall": 0.9485861182519281,
"support": 27619.0
},
"step": 486
},
{
"epoch": 6.17,
"grad_norm": 4.767655849456787,
"learning_rate": 1.7530864197530865e-05,
"loss": 0.1585,
"step": 500
},
{
"epoch": 7.0,
"eval_B": {
"f1-score": 0.8802588996763754,
"precision": 0.85,
"recall": 0.912751677852349,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.958423523458803,
"precision": 0.9354661691269275,
"recall": 0.9825360230547551,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9139526911210146,
"precision": 0.9664088931851136,
"recall": 0.8668978972469109,
"support": 9226.0
},
"eval_accuracy": 0.9412723125384699,
"eval_loss": 0.2591131329536438,
"eval_macro avg": {
"f1-score": 0.9175450380853977,
"precision": 0.917291687437347,
"recall": 0.9207285327180049,
"support": 27619.0
},
"eval_runtime": 4.617,
"eval_samples_per_second": 17.327,
"eval_steps_per_second": 2.166,
"eval_weighted avg": {
"f1-score": 0.9406164485555297,
"precision": 0.9425749115781907,
"recall": 0.9412723125384699,
"support": 27619.0
},
"step": 567
},
{
"epoch": 8.0,
"eval_B": {
"f1-score": 0.8687910028116214,
"precision": 0.84967919340055,
"recall": 0.8887823585810163,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9595010203788336,
"precision": 0.956997878562009,
"recall": 0.9620172910662824,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9212035166275324,
"precision": 0.9282491471332673,
"recall": 0.9142640364188164,
"support": 9226.0
},
"eval_accuracy": 0.9432999022412107,
"eval_loss": 0.2693409323692322,
"eval_macro avg": {
"f1-score": 0.9164985132726624,
"precision": 0.9116420730319422,
"recall": 0.9216878953553717,
"support": 27619.0
},
"eval_runtime": 4.6153,
"eval_samples_per_second": 17.333,
"eval_steps_per_second": 2.167,
"eval_weighted avg": {
"f1-score": 0.9432823550422136,
"precision": 0.9433417293609165,
"recall": 0.9432999022412107,
"support": 27619.0
},
"step": 648
},
{
"epoch": 9.0,
"eval_B": {
"f1-score": 0.8645343367826904,
"precision": 0.8485687903970452,
"recall": 0.8811121764141898,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9567003348057282,
"precision": 0.934535260814599,
"recall": 0.9799423631123919,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9104673003585861,
"precision": 0.9586479683567062,
"recall": 0.8668978972469109,
"support": 9226.0
},
"eval_accuracy": 0.9384481697382236,
"eval_loss": 0.3087053894996643,
"eval_macro avg": {
"f1-score": 0.910567323982335,
"precision": 0.9139173398561168,
"recall": 0.9093174789244975,
"support": 27619.0
},
"eval_runtime": 4.6156,
"eval_samples_per_second": 17.333,
"eval_steps_per_second": 2.167,
"eval_weighted avg": {
"f1-score": 0.9377758584761231,
"precision": 0.9393435743356523,
"recall": 0.9384481697382236,
"support": 27619.0
},
"step": 729
},
{
"epoch": 10.0,
"eval_B": {
"f1-score": 0.8803126526624329,
"precision": 0.897410358565737,
"recall": 0.8638542665388304,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9590983771942466,
"precision": 0.9592919333448654,
"recall": 0.9589048991354466,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9220456265542221,
"precision": 0.9197584124245038,
"recall": 0.9243442445263386,
"support": 9226.0
},
"eval_accuracy": 0.9437705927079184,
"eval_loss": 0.2796257436275482,
"eval_macro avg": {
"f1-score": 0.9204855521369671,
"precision": 0.9254869014450354,
"recall": 0.9157011367335386,
"support": 27619.0
},
"eval_runtime": 4.641,
"eval_samples_per_second": 17.238,
"eval_steps_per_second": 2.155,
"eval_weighted avg": {
"f1-score": 0.9437458232244597,
"precision": 0.9437490553802075,
"recall": 0.9437705927079184,
"support": 27619.0
},
"step": 810
},
{
"epoch": 11.0,
"eval_B": {
"f1-score": 0.8830917874396135,
"precision": 0.8899707887049659,
"recall": 0.8763183125599233,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9574634882843123,
"precision": 0.9647182727751448,
"recall": 0.9503170028818444,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9200619426496501,
"precision": 0.9067466582465004,
"recall": 0.9337741166269239,
"support": 9226.0
},
"eval_accuracy": 0.9419964517180202,
"eval_loss": 0.3198166787624359,
"eval_macro avg": {
"f1-score": 0.9202057394578587,
"precision": 0.9204785732422037,
"recall": 0.9201364773562305,
"support": 27619.0
},
"eval_runtime": 4.6345,
"eval_samples_per_second": 17.262,
"eval_steps_per_second": 2.158,
"eval_weighted avg": {
"f1-score": 0.942161111514465,
"precision": 0.9425303680165918,
"recall": 0.9419964517180202,
"support": 27619.0
},
"step": 891
},
{
"epoch": 12.0,
"eval_B": {
"f1-score": 0.871306005719733,
"precision": 0.8663507109004739,
"recall": 0.8763183125599233,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9576855173984852,
"precision": 0.9464205312922107,
"recall": 0.9692219020172911,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9162135168127844,
"precision": 0.9386084583901774,
"recall": 0.8948623455451984,
"support": 9226.0
},
"eval_accuracy": 0.9408740359897172,
"eval_loss": 0.35264310240745544,
"eval_macro avg": {
"f1-score": 0.9150683466436677,
"precision": 0.917126566860954,
"recall": 0.9134675200408043,
"support": 27619.0
},
"eval_runtime": 4.6326,
"eval_samples_per_second": 17.269,
"eval_steps_per_second": 2.159,
"eval_weighted avg": {
"f1-score": 0.9405699625961891,
"precision": 0.9407871989028143,
"recall": 0.9408740359897172,
"support": 27619.0
},
"step": 972
},
{
"epoch": 12.35,
"grad_norm": 0.3602512776851654,
"learning_rate": 1.506172839506173e-05,
"loss": 0.021,
"step": 1000
},
{
"epoch": 13.0,
"eval_B": {
"f1-score": 0.877092300334768,
"precision": 0.875,
"recall": 0.8791946308724832,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9597242793665263,
"precision": 0.9486991778353419,
"recall": 0.9710086455331413,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9203392649259936,
"precision": 0.9419040054464994,
"recall": 0.8997398655972252,
"support": 9226.0
},
"eval_accuracy": 0.943734385748941,
"eval_loss": 0.3594338297843933,
"eval_macro avg": {
"f1-score": 0.9190519482090961,
"precision": 0.9218677277606138,
"recall": 0.9166477140009498,
"support": 27619.0
},
"eval_runtime": 4.6042,
"eval_samples_per_second": 17.376,
"eval_steps_per_second": 2.172,
"eval_weighted avg": {
"f1-score": 0.943447393984779,
"precision": 0.9436461164304495,
"recall": 0.943734385748941,
"support": 27619.0
},
"step": 1053
},
{
"epoch": 14.0,
"eval_B": {
"f1-score": 0.8754052802223252,
"precision": 0.8467741935483871,
"recall": 0.9060402684563759,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9574973153224439,
"precision": 0.9392880904856953,
"recall": 0.9764265129682997,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9130164471825015,
"precision": 0.9539388213062477,
"recall": 0.8754606546715803,
"support": 9226.0
},
"eval_accuracy": 0.9400412759332344,
"eval_loss": 0.3840242028236389,
"eval_macro avg": {
"f1-score": 0.9153063475757568,
"precision": 0.9133337017801101,
"recall": 0.9193091453654186,
"support": 27619.0
},
"eval_runtime": 4.6182,
"eval_samples_per_second": 17.323,
"eval_steps_per_second": 2.165,
"eval_weighted avg": {
"f1-score": 0.9395385738014427,
"precision": 0.9406884180878825,
"recall": 0.9400412759332344,
"support": 27619.0
},
"step": 1134
},
{
"epoch": 15.0,
"eval_B": {
"f1-score": 0.8797709923664121,
"precision": 0.8755935422602089,
"recall": 0.8839884947267498,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9603688219460511,
"precision": 0.954302299112224,
"recall": 0.9665129682997118,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.922722283205269,
"precision": 0.9346230820547031,
"recall": 0.9111207457186213,
"support": 9226.0
},
"eval_accuracy": 0.9448930084362215,
"eval_loss": 0.3762665390968323,
"eval_macro avg": {
"f1-score": 0.9209540325059108,
"precision": 0.9215063078090453,
"recall": 0.920540736248361,
"support": 27619.0
},
"eval_runtime": 4.586,
"eval_samples_per_second": 17.444,
"eval_steps_per_second": 2.181,
"eval_weighted avg": {
"f1-score": 0.944749483712443,
"precision": 0.9447562007752335,
"recall": 0.9448930084362215,
"support": 27619.0
},
"step": 1215
},
{
"epoch": 16.0,
"eval_B": {
"f1-score": 0.8858669256920835,
"precision": 0.8976377952755905,
"recall": 0.8744007670182167,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9586212929752427,
"precision": 0.9701351590627397,
"recall": 0.9473775216138328,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9235412474849094,
"precision": 0.9027950310559006,
"recall": 0.9452633860828095,
"support": 9226.0
},
"eval_accuracy": 0.9439154205438285,
"eval_loss": 0.36898088455200195,
"eval_macro avg": {
"f1-score": 0.9226764887174118,
"precision": 0.923522661798077,
"recall": 0.922347224904953,
"support": 27619.0
},
"eval_runtime": 4.6024,
"eval_samples_per_second": 17.382,
"eval_steps_per_second": 2.173,
"eval_weighted avg": {
"f1-score": 0.9441554794131966,
"precision": 0.9449027186622512,
"recall": 0.9439154205438285,
"support": 27619.0
},
"step": 1296
},
{
"epoch": 17.0,
"eval_B": {
"f1-score": 0.8863309352517986,
"precision": 0.8867562380038387,
"recall": 0.8859060402684564,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9630614486386496,
"precision": 0.9659090909090909,
"recall": 0.9602305475504322,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9296685529506872,
"precision": 0.9245363918962375,
"recall": 0.9348580099718188,
"support": 9226.0
},
"eval_accuracy": 0.9489481878417032,
"eval_loss": 0.3569599986076355,
"eval_macro avg": {
"f1-score": 0.9263536456137119,
"precision": 0.9257339069363891,
"recall": 0.9269981992635691,
"support": 27619.0
},
"eval_runtime": 4.6143,
"eval_samples_per_second": 17.337,
"eval_steps_per_second": 2.167,
"eval_weighted avg": {
"f1-score": 0.9490090650954502,
"precision": 0.9490996138580476,
"recall": 0.9489481878417032,
"support": 27619.0
},
"step": 1377
},
{
"epoch": 18.0,
"eval_B": {
"f1-score": 0.8778409090909091,
"precision": 0.8671655753040225,
"recall": 0.8887823585810163,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9557028055829692,
"precision": 0.9355230471984544,
"recall": 0.9767723342939482,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9098012570069645,
"precision": 0.9524599881446354,
"recall": 0.8707999132885325,
"support": 9226.0
},
"eval_accuracy": 0.938049893189471,
"eval_loss": 0.4151785373687744,
"eval_macro avg": {
"f1-score": 0.9144483238936143,
"precision": 0.9183828702157042,
"recall": 0.9121182020544989,
"support": 27619.0
},
"eval_runtime": 4.6304,
"eval_samples_per_second": 17.277,
"eval_steps_per_second": 2.16,
"eval_weighted avg": {
"f1-score": 0.9374292386470396,
"precision": 0.9385993125948691,
"recall": 0.938049893189471,
"support": 27619.0
},
"step": 1458
},
{
"epoch": 18.52,
"grad_norm": 10.955327033996582,
"learning_rate": 1.2592592592592593e-05,
"loss": 0.0058,
"step": 1500
},
{
"epoch": 19.0,
"eval_B": {
"f1-score": 0.8850241545893719,
"precision": 0.8919182083739046,
"recall": 0.87823585810163,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9626489459211732,
"precision": 0.956838628857761,
"recall": 0.9685302593659942,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.927475898334794,
"precision": 0.9375415282392027,
"recall": 0.9176241057879905,
"support": 9226.0
},
"eval_accuracy": 0.9481154277852203,
"eval_loss": 0.3719603419303894,
"eval_macro avg": {
"f1-score": 0.9250496662817796,
"precision": 0.9287661218236227,
"recall": 0.9214634077518715,
"support": 27619.0
},
"eval_runtime": 4.6041,
"eval_samples_per_second": 17.376,
"eval_steps_per_second": 2.172,
"eval_weighted avg": {
"f1-score": 0.9479681394332119,
"precision": 0.9479408755404257,
"recall": 0.9481154277852203,
"support": 27619.0
},
"step": 1539
},
{
"epoch": 20.0,
"eval_B": {
"f1-score": 0.8742857142857142,
"precision": 0.8684957426679281,
"recall": 0.8801534036433365,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9602204921293401,
"precision": 0.9469289396996189,
"recall": 0.9738904899135447,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9205305394560855,
"precision": 0.9473503097040605,
"recall": 0.8951875135486668,
"support": 9226.0
},
"eval_accuracy": 0.9440602483797386,
"eval_loss": 0.36765894293785095,
"eval_macro avg": {
"f1-score": 0.9183455819570466,
"precision": 0.9209249973572025,
"recall": 0.9164104690351826,
"support": 27619.0
},
"eval_runtime": 4.6404,
"eval_samples_per_second": 17.24,
"eval_steps_per_second": 2.155,
"eval_weighted avg": {
"f1-score": 0.9437170171065533,
"precision": 0.9441077562808465,
"recall": 0.9440602483797386,
"support": 27619.0
},
"step": 1620
},
{
"epoch": 21.0,
"eval_B": {
"f1-score": 0.8885767790262172,
"precision": 0.868252516010979,
"recall": 0.909875359539789,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9616143138880155,
"precision": 0.9537903271531439,
"recall": 0.9695677233429395,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9245376759591498,
"precision": 0.9420632242096973,
"recall": 0.9076522870149577,
"support": 9226.0
},
"eval_accuracy": 0.9466309424671422,
"eval_loss": 0.374899685382843,
"eval_macro avg": {
"f1-score": 0.9249095896244609,
"precision": 0.9213686891246068,
"recall": 0.9290317899658954,
"support": 27619.0
},
"eval_runtime": 4.5979,
"eval_samples_per_second": 17.399,
"eval_steps_per_second": 2.175,
"eval_weighted avg": {
"f1-score": 0.9464708542988716,
"precision": 0.9466427045463329,
"recall": 0.9466309424671422,
"support": 27619.0
},
"step": 1701
},
{
"epoch": 22.0,
"eval_B": {
"f1-score": 0.875898418782942,
"precision": 0.8754789272030651,
"recall": 0.8763183125599233,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9606979453806588,
"precision": 0.9475306911822412,
"recall": 0.9742363112391931,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9217236387930074,
"precision": 0.94757326007326,
"recall": 0.897246910903967,
"support": 9226.0
},
"eval_accuracy": 0.9448205945182664,
"eval_loss": 0.3899074196815491,
"eval_macro avg": {
"f1-score": 0.919440000985536,
"precision": 0.9235276261528554,
"recall": 0.9159338449010278,
"support": 27619.0
},
"eval_runtime": 4.5958,
"eval_samples_per_second": 17.407,
"eval_steps_per_second": 2.176,
"eval_weighted avg": {
"f1-score": 0.9444764001104068,
"precision": 0.9448239585256736,
"recall": 0.9448205945182664,
"support": 27619.0
},
"step": 1782
},
{
"epoch": 23.0,
"eval_B": {
"f1-score": 0.8766079085278704,
"precision": 0.8712121212121212,
"recall": 0.8820709491850431,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9571775605897973,
"precision": 0.9385663638378019,
"recall": 0.9765417867435159,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9132322264193494,
"precision": 0.9515920573375631,
"recall": 0.877845220030349,
"support": 9226.0
},
"eval_accuracy": 0.9400050689742568,
"eval_loss": 0.4506397247314453,
"eval_macro avg": {
"f1-score": 0.9156725651790056,
"precision": 0.9204568474624955,
"recall": 0.9121526519863027,
"support": 27619.0
},
"eval_runtime": 4.6193,
"eval_samples_per_second": 17.319,
"eval_steps_per_second": 2.165,
"eval_weighted avg": {
"f1-score": 0.939455202786939,
"precision": 0.9403739808105457,
"recall": 0.9400050689742568,
"support": 27619.0
},
"step": 1863
},
{
"epoch": 24.0,
"eval_B": {
"f1-score": 0.8786853552440792,
"precision": 0.8859649122807017,
"recall": 0.8715244487056567,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9558590028580501,
"precision": 0.9575452599919023,
"recall": 0.954178674351585,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9160280626011873,
"precision": 0.9121883061049011,
"recall": 0.9199002818122697,
"support": 9226.0
},
"eval_accuracy": 0.9396067924255042,
"eval_loss": 0.42392534017562866,
"eval_macro avg": {
"f1-score": 0.9168574735677723,
"precision": 0.9185661594591683,
"recall": 0.9152011349565038,
"support": 27619.0
},
"eval_runtime": 4.6242,
"eval_samples_per_second": 17.3,
"eval_steps_per_second": 2.163,
"eval_weighted avg": {
"f1-score": 0.9396392856607877,
"precision": 0.9396908279261412,
"recall": 0.9396067924255042,
"support": 27619.0
},
"step": 1944
},
{
"epoch": 24.69,
"grad_norm": 0.030169092118740082,
"learning_rate": 1.0123456790123458e-05,
"loss": 0.0026,
"step": 2000
},
{
"epoch": 25.0,
"eval_B": {
"f1-score": 0.8887832699619772,
"precision": 0.88124410933082,
"recall": 0.8964525407478428,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9607269808508199,
"precision": 0.9556860955857192,
"recall": 0.9658213256484149,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9237260273972602,
"precision": 0.9340647163120568,
"recall": 0.9136137004118795,
"support": 9226.0
},
"eval_accuracy": 0.9457619754516818,
"eval_loss": 0.398232638835907,
"eval_macro avg": {
"f1-score": 0.9244120927366858,
"precision": 0.9236649737428652,
"recall": 0.9252958556027124,
"support": 27619.0
},
"eval_runtime": 4.5953,
"eval_samples_per_second": 17.409,
"eval_steps_per_second": 2.176,
"eval_weighted avg": {
"f1-score": 0.9456501103261954,
"precision": 0.9456523566073828,
"recall": 0.9457619754516818,
"support": 27619.0
},
"step": 2025
},
{
"epoch": 26.0,
"eval_B": {
"f1-score": 0.8827717133364974,
"precision": 0.8740601503759399,
"recall": 0.8916586768935763,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9590283123770422,
"precision": 0.9489364103142809,
"recall": 0.9693371757925072,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9188171447557869,
"precision": 0.9393115942028986,
"recall": 0.8991979189247779,
"support": 9226.0
},
"eval_accuracy": 0.9429740396104132,
"eval_loss": 0.4066696763038635,
"eval_macro avg": {
"f1-score": 0.9202057234897755,
"precision": 0.9207693849643731,
"recall": 0.9200645905369539,
"support": 27619.0
},
"eval_runtime": 4.6361,
"eval_samples_per_second": 17.256,
"eval_steps_per_second": 2.157,
"eval_weighted avg": {
"f1-score": 0.9427162132687112,
"precision": 0.942893668268613,
"recall": 0.9429740396104132,
"support": 27619.0
},
"step": 2106
},
{
"epoch": 27.0,
"eval_B": {
"f1-score": 0.8731307284129282,
"precision": 0.8786407766990292,
"recall": 0.8676893576222435,
"support": 1043.0
},
"eval_I": {
"f1-score": 0.9568967963451661,
"precision": 0.9424291543234028,
"recall": 0.9718155619596541,
"support": 17350.0
},
"eval_O": {
"f1-score": 0.9140816781968311,
"precision": 0.9418257070590941,
"recall": 0.8879254281378712,
"support": 9226.0
},
"eval_accuracy": 0.9398602411383468,
"eval_loss": 0.44146138429641724,
"eval_macro avg": {
"f1-score": 0.9147030676516418,
"precision": 0.920965212693842,
"recall": 0.9091434492399229,
"support": 27619.0
},
"eval_runtime": 4.6584,
"eval_samples_per_second": 17.173,
"eval_steps_per_second": 2.147,
"eval_weighted avg": {
"f1-score": 0.9394312730137688,
"precision": 0.9398186802902107,
"recall": 0.9398602411383468,
"support": 27619.0
},
"step": 2187
}
],
"logging_steps": 500,
"max_steps": 4050,
"num_input_tokens_seen": 0,
"num_train_epochs": 50,
"save_steps": 500,
"total_flos": 3882239623861200.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}