longformer-simple / checkpoint-4050 /trainer_state.json
Theoreticallyhugo's picture
Training in progress, epoch 50, checkpoint
a5118e5 verified
raw
history blame
No virus
67.5 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 50.0,
"eval_steps": 500,
"global_step": 4050,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_Claim": {
"f1-score": 0.4888068880688806,
"precision": 0.6228840125391849,
"recall": 0.40222672064777326,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.7325278370054489,
"precision": 0.7604525332021643,
"recall": 0.706581352833638,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.8634848138844484,
"precision": 0.827613377692173,
"recall": 0.902606702950444,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8792819197655248,
"precision": 0.8543458914268763,
"recall": 0.9057173407132524,
"support": 15899.0
},
"eval_accuracy": 0.8174925373134329,
"eval_loss": 0.49806877970695496,
"eval_macro avg": {
"f1-score": 0.7410253646810757,
"precision": 0.7663239537150996,
"recall": 0.7292830292862769,
"support": 33500.0
},
"eval_runtime": 4.9194,
"eval_samples_per_second": 16.465,
"eval_steps_per_second": 2.236,
"eval_weighted avg": {
"f1-score": 0.8071778099281224,
"precision": 0.80572410140167,
"recall": 0.8174925373134329,
"support": 33500.0
},
"step": 81
},
{
"epoch": 2.0,
"eval_Claim": {
"f1-score": 0.5993816254416962,
"precision": 0.6593780369290574,
"recall": 0.5493927125506073,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8017817371937639,
"precision": 0.7197382769901853,
"recall": 0.9049360146252285,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9058997050147494,
"precision": 0.9337184554575859,
"recall": 0.8796906330564308,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8854737486606459,
"precision": 0.862579028987236,
"recall": 0.9096169570413234,
"support": 15899.0
},
"eval_accuracy": 0.8468358208955223,
"eval_loss": 0.4562840759754181,
"eval_macro avg": {
"f1-score": 0.7981342040777137,
"precision": 0.7938534495910161,
"recall": 0.8109090793183975,
"support": 33500.0
},
"eval_runtime": 4.904,
"eval_samples_per_second": 16.517,
"eval_steps_per_second": 2.243,
"eval_weighted avg": {
"f1-score": 0.8442053257085675,
"precision": 0.8455251408465503,
"recall": 0.8468358208955223,
"support": 33500.0
},
"step": 162
},
{
"epoch": 3.0,
"eval_Claim": {
"f1-score": 0.6235462150581514,
"precision": 0.6285479226655697,
"recall": 0.6186234817813765,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8338848444738584,
"precision": 0.8059701492537313,
"recall": 0.8638025594149908,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9066250826953974,
"precision": 0.8974646833193002,
"recall": 0.9159744103886184,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8827730692315017,
"precision": 0.8911176621379133,
"recall": 0.8745833071262343,
"support": 15899.0
},
"eval_accuracy": 0.8490746268656716,
"eval_loss": 0.43820467591285706,
"eval_macro avg": {
"f1-score": 0.8117073028647273,
"precision": 0.8057751043441286,
"recall": 0.818245939677805,
"support": 33500.0
},
"eval_runtime": 4.9179,
"eval_samples_per_second": 16.47,
"eval_steps_per_second": 2.237,
"eval_weighted avg": {
"f1-score": 0.8488105033097496,
"precision": 0.8488213959184715,
"recall": 0.8490746268656716,
"support": 33500.0
},
"step": 243
},
{
"epoch": 4.0,
"eval_Claim": {
"f1-score": 0.6415287144900879,
"precision": 0.6477507222451506,
"recall": 0.6354251012145749,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8005126014523707,
"precision": 0.7514033680834001,
"recall": 0.856489945155393,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9065169636319258,
"precision": 0.9273871354374751,
"recall": 0.8865654540246348,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8932505382719131,
"precision": 0.8863636363636364,
"recall": 0.9002452984464432,
"support": 15899.0
},
"eval_accuracy": 0.8540597014925373,
"eval_loss": 0.4749625325202942,
"eval_macro avg": {
"f1-score": 0.8104522044615743,
"precision": 0.8032262155324155,
"recall": 0.8196814497102615,
"support": 33500.0
},
"eval_runtime": 4.8972,
"eval_samples_per_second": 16.54,
"eval_steps_per_second": 2.246,
"eval_weighted avg": {
"f1-score": 0.8542213698406307,
"precision": 0.8551874645146165,
"recall": 0.8540597014925373,
"support": 33500.0
},
"step": 324
},
{
"epoch": 5.0,
"eval_Claim": {
"f1-score": 0.6099687080911935,
"precision": 0.6808882235528942,
"recall": 0.5524291497975709,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8315018315018317,
"precision": 0.8330275229357799,
"recall": 0.829981718464351,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9103454839630722,
"precision": 0.9074091642159188,
"recall": 0.9133008689009835,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8913988368533824,
"precision": 0.8682249120505635,
"recall": 0.915843763758727,
"support": 15899.0
},
"eval_accuracy": 0.8558507462686568,
"eval_loss": 0.5162495374679565,
"eval_macro avg": {
"f1-score": 0.8108037151023699,
"precision": 0.8223874556887891,
"recall": 0.8028888752304081,
"support": 33500.0
},
"eval_runtime": 4.9413,
"eval_samples_per_second": 16.392,
"eval_steps_per_second": 2.226,
"eval_weighted avg": {
"f1-score": 0.8519095458499607,
"precision": 0.8505509283002989,
"recall": 0.8558507462686568,
"support": 33500.0
},
"step": 405
},
{
"epoch": 6.0,
"eval_Claim": {
"f1-score": 0.646279703934554,
"precision": 0.6227477477477478,
"recall": 0.6716599190283401,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8414067684140676,
"precision": 0.8152593227603944,
"recall": 0.8692870201096892,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9112506693929215,
"precision": 0.9295788637266588,
"recall": 0.8936312422419556,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8861383012314494,
"precision": 0.8897343224906474,
"recall": 0.8825712308950249,
"support": 15899.0
},
"eval_accuracy": 0.8540597014925373,
"eval_loss": 0.552895724773407,
"eval_macro avg": {
"f1-score": 0.8212688607432481,
"precision": 0.8143300641813621,
"recall": 0.8292873530687525,
"support": 33500.0
},
"eval_runtime": 4.9336,
"eval_samples_per_second": 16.418,
"eval_steps_per_second": 2.23,
"eval_weighted avg": {
"f1-score": 0.8556973390614198,
"precision": 0.8579560210496036,
"recall": 0.8540597014925373,
"support": 33500.0
},
"step": 486
},
{
"epoch": 6.17,
"grad_norm": 9.790245056152344,
"learning_rate": 1.7530864197530865e-05,
"loss": 0.3328,
"step": 500
},
{
"epoch": 7.0,
"eval_Claim": {
"f1-score": 0.6202600808654791,
"precision": 0.6739491807171694,
"recall": 0.5744939271255061,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8670041417395307,
"precision": 0.8730305838739574,
"recall": 0.8610603290676416,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9109892729439809,
"precision": 0.9097314797181489,
"recall": 0.9122505490308412,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8872356123954747,
"precision": 0.8677611401767996,
"recall": 0.9076042518397384,
"support": 15899.0
},
"eval_accuracy": 0.8568955223880597,
"eval_loss": 0.5894299149513245,
"eval_macro avg": {
"f1-score": 0.8213722769861164,
"precision": 0.8311180961215189,
"recall": 0.8138522642659318,
"support": 33500.0
},
"eval_runtime": 4.9037,
"eval_samples_per_second": 16.518,
"eval_steps_per_second": 2.243,
"eval_weighted avg": {
"f1-score": 0.8539713289140156,
"precision": 0.8526463291050193,
"recall": 0.8568955223880597,
"support": 33500.0
},
"step": 567
},
{
"epoch": 8.0,
"eval_Claim": {
"f1-score": 0.6441500279902966,
"precision": 0.5974385600553825,
"recall": 0.6987854251012146,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.834510811389424,
"precision": 0.7849375755134917,
"recall": 0.8907678244972578,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9059161060546101,
"precision": 0.9398542543364466,
"recall": 0.874343550081161,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8801401497053671,
"precision": 0.8915849251419721,
"recall": 0.8689854707843261,
"support": 15899.0
},
"eval_accuracy": 0.8469850746268657,
"eval_loss": 0.5878348350524902,
"eval_macro avg": {
"f1-score": 0.8161792737849245,
"precision": 0.8034538287618231,
"recall": 0.8332205676159898,
"support": 33500.0
},
"eval_runtime": 4.9086,
"eval_samples_per_second": 16.502,
"eval_steps_per_second": 2.241,
"eval_weighted avg": {
"f1-score": 0.8504184600736624,
"precision": 0.8563340964894008,
"recall": 0.8469850746268657,
"support": 33500.0
},
"step": 648
},
{
"epoch": 9.0,
"eval_Claim": {
"f1-score": 0.6323094193285029,
"precision": 0.6375797489195307,
"recall": 0.6271255060728745,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8525914292124747,
"precision": 0.8373732921992068,
"recall": 0.8683729433272395,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9054132151801194,
"precision": 0.9134194927606647,
"recall": 0.8975460708488494,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8888055034396498,
"precision": 0.883775884584292,
"recall": 0.8938926976539405,
"support": 15899.0
},
"eval_accuracy": 0.8540298507462687,
"eval_loss": 0.7043101191520691,
"eval_macro avg": {
"f1-score": 0.8197798917901866,
"precision": 0.8180371046159236,
"recall": 0.821734304475726,
"support": 33500.0
},
"eval_runtime": 4.935,
"eval_samples_per_second": 16.414,
"eval_steps_per_second": 2.229,
"eval_weighted avg": {
"f1-score": 0.853808653145316,
"precision": 0.8537078167069386,
"recall": 0.8540298507462687,
"support": 33500.0
},
"step": 729
},
{
"epoch": 10.0,
"eval_Claim": {
"f1-score": 0.6347110795166041,
"precision": 0.6368453230079478,
"recall": 0.6325910931174089,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8442090055381651,
"precision": 0.8921119592875318,
"recall": 0.8011882998171846,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9106677408653618,
"precision": 0.9040270982310877,
"recall": 0.9174066647569942,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8841656478259506,
"precision": 0.881375,
"recall": 0.8869740235234921,
"support": 15899.0
},
"eval_accuracy": 0.8533731343283583,
"eval_loss": 0.7164433002471924,
"eval_macro avg": {
"f1-score": 0.8184383684365204,
"precision": 0.8285898451316418,
"recall": 0.8095400203037699,
"support": 33500.0
},
"eval_runtime": 4.9093,
"eval_samples_per_second": 16.499,
"eval_steps_per_second": 2.241,
"eval_weighted avg": {
"f1-score": 0.8530559678148433,
"precision": 0.8530989190255094,
"recall": 0.8533731343283583,
"support": 33500.0
},
"step": 810
},
{
"epoch": 11.0,
"eval_Claim": {
"f1-score": 0.645658668208538,
"precision": 0.6161486113665624,
"recall": 0.6781376518218624,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8331177231565331,
"precision": 0.7885714285714286,
"recall": 0.8829981718464351,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9059711059612445,
"precision": 0.9366843393148451,
"recall": 0.8772080588179128,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8902977542265961,
"precision": 0.8929452704840241,
"recall": 0.8876658909365369,
"support": 15899.0
},
"eval_accuracy": 0.8531940298507462,
"eval_loss": 0.7852362394332886,
"eval_macro avg": {
"f1-score": 0.818761312888228,
"precision": 0.8085874124342151,
"recall": 0.8315024433556868,
"support": 33500.0
},
"eval_runtime": 4.89,
"eval_samples_per_second": 16.565,
"eval_steps_per_second": 2.25,
"eval_weighted avg": {
"f1-score": 0.8553879040715653,
"precision": 0.8589850855801485,
"recall": 0.8531940298507462,
"support": 33500.0
},
"step": 891
},
{
"epoch": 12.0,
"eval_Claim": {
"f1-score": 0.6329215627196947,
"precision": 0.6280645804265498,
"recall": 0.6378542510121458,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8528368794326241,
"precision": 0.8278829604130808,
"recall": 0.8793418647166362,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9126755145220234,
"precision": 0.9193064031773709,
"recall": 0.9061395970591044,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8876634630534112,
"precision": 0.8894291487749432,
"recall": 0.88590477388515,
"support": 15899.0
},
"eval_accuracy": 0.8552238805970149,
"eval_loss": 0.8324545621871948,
"eval_macro avg": {
"f1-score": 0.8215243549319383,
"precision": 0.8161707731979861,
"recall": 0.827310121668259,
"support": 33500.0
},
"eval_runtime": 4.8939,
"eval_samples_per_second": 16.551,
"eval_steps_per_second": 2.248,
"eval_weighted avg": {
"f1-score": 0.8556433335734092,
"precision": 0.8562082669116539,
"recall": 0.8552238805970149,
"support": 33500.0
},
"step": 972
},
{
"epoch": 12.35,
"grad_norm": 2.6090729236602783,
"learning_rate": 1.506172839506173e-05,
"loss": 0.0589,
"step": 1000
},
{
"epoch": 13.0,
"eval_Claim": {
"f1-score": 0.6431152147731835,
"precision": 0.6377388535031847,
"recall": 0.648582995951417,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8326953388876379,
"precision": 0.8206835330670218,
"recall": 0.8450639853747715,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.904438383337466,
"precision": 0.9408790755262072,
"recall": 0.8707151723479424,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.892013567684243,
"precision": 0.8749621922448733,
"recall": 0.9097427511164224,
"support": 15899.0
},
"eval_accuracy": 0.8548059701492537,
"eval_loss": 0.8543176651000977,
"eval_macro avg": {
"f1-score": 0.8180656261706326,
"precision": 0.8185659135853218,
"recall": 0.8185262261976383,
"support": 33500.0
},
"eval_runtime": 4.9338,
"eval_samples_per_second": 16.417,
"eval_steps_per_second": 2.23,
"eval_weighted avg": {
"f1-score": 0.8553204019036041,
"precision": 0.8570428644520474,
"recall": 0.8548059701492537,
"support": 33500.0
},
"step": 1053
},
{
"epoch": 14.0,
"eval_Claim": {
"f1-score": 0.6541843971631205,
"precision": 0.6138420585625555,
"recall": 0.7002024291497976,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8336528221512247,
"precision": 0.7806142800159553,
"recall": 0.8944241316270567,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9079845651528643,
"precision": 0.9421004003695719,
"recall": 0.8762532225723288,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.889389516436096,
"precision": 0.8974194787731319,
"recall": 0.8815019812566828,
"support": 15899.0
},
"eval_accuracy": 0.8539701492537314,
"eval_loss": 0.9095103144645691,
"eval_macro avg": {
"f1-score": 0.8213028252258264,
"precision": 0.8084940544303038,
"recall": 0.8380954411514664,
"support": 33500.0
},
"eval_runtime": 4.8917,
"eval_samples_per_second": 16.559,
"eval_steps_per_second": 2.249,
"eval_weighted avg": {
"f1-score": 0.8568784946124219,
"precision": 0.8619418985092084,
"recall": 0.8539701492537314,
"support": 33500.0
},
"step": 1134
},
{
"epoch": 15.0,
"eval_Claim": {
"f1-score": 0.631984585741811,
"precision": 0.6332046332046332,
"recall": 0.6307692307692307,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8264781491002571,
"precision": 0.7778225806451613,
"recall": 0.8816270566727605,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.8970963775654971,
"precision": 0.8886910896655111,
"recall": 0.9056621789363124,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8838946528332003,
"precision": 0.8974458706080644,
"recall": 0.8707465878357129,
"support": 15899.0
},
"eval_accuracy": 0.8469850746268657,
"eval_loss": 0.9576827883720398,
"eval_macro avg": {
"f1-score": 0.8098634413101913,
"precision": 0.7992910435308425,
"recall": 0.8222012635535042,
"support": 33500.0
},
"eval_runtime": 4.8883,
"eval_samples_per_second": 16.57,
"eval_steps_per_second": 2.25,
"eval_weighted avg": {
"f1-score": 0.8471244624308779,
"precision": 0.8479301603984184,
"recall": 0.8469850746268657,
"support": 33500.0
},
"step": 1215
},
{
"epoch": 16.0,
"eval_Claim": {
"f1-score": 0.6332942805410586,
"precision": 0.6682400539447066,
"recall": 0.6018218623481781,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8478211270819075,
"precision": 0.8464692482915718,
"recall": 0.8491773308957953,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9089773764349872,
"precision": 0.9145563502803016,
"recall": 0.9034660555714695,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8924681415656145,
"precision": 0.8759539672925499,
"recall": 0.9096169570413234,
"support": 15899.0
},
"eval_accuracy": 0.8583582089552239,
"eval_loss": 0.9394893646240234,
"eval_macro avg": {
"f1-score": 0.8206402314058919,
"precision": 0.8263049049522826,
"recall": 0.8160205514641916,
"support": 33500.0
},
"eval_runtime": 4.896,
"eval_samples_per_second": 16.544,
"eval_steps_per_second": 2.247,
"eval_weighted avg": {
"f1-score": 0.8564948483009961,
"precision": 0.8554663093796615,
"recall": 0.8583582089552239,
"support": 33500.0
},
"step": 1296
},
{
"epoch": 17.0,
"eval_Claim": {
"f1-score": 0.6431561322729175,
"precision": 0.6657279029462738,
"recall": 0.6220647773279352,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8454073572556984,
"precision": 0.8350423539901917,
"recall": 0.8560329067641682,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.909496556407023,
"precision": 0.9241991128634796,
"recall": 0.8952544638594481,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8953850902917116,
"precision": 0.8791828322017459,
"recall": 0.9121957355808541,
"support": 15899.0
},
"eval_accuracy": 0.8604477611940299,
"eval_loss": 0.9581068754196167,
"eval_macro avg": {
"f1-score": 0.8233612840568376,
"precision": 0.8260380505004228,
"recall": 0.8213869708831014,
"support": 33500.0
},
"eval_runtime": 4.8973,
"eval_samples_per_second": 16.54,
"eval_steps_per_second": 2.246,
"eval_weighted avg": {
"f1-score": 0.859338136623951,
"precision": 0.8588965274411914,
"recall": 0.8604477611940299,
"support": 33500.0
},
"step": 1377
},
{
"epoch": 18.0,
"eval_Claim": {
"f1-score": 0.6493941483597675,
"precision": 0.632508155824218,
"recall": 0.6672064777327935,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8654810761513908,
"precision": 0.8635122838944495,
"recall": 0.8674588665447898,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9125304729400293,
"precision": 0.9323503038756601,
"recall": 0.8935357586173971,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8928739085531874,
"precision": 0.8885635978572318,
"recall": 0.8972262406440656,
"support": 15899.0
},
"eval_accuracy": 0.8602089552238806,
"eval_loss": 0.9848275184631348,
"eval_macro avg": {
"f1-score": 0.8300699015010937,
"precision": 0.8292335853628898,
"recall": 0.8313568358847614,
"support": 33500.0
},
"eval_runtime": 4.9301,
"eval_samples_per_second": 16.43,
"eval_steps_per_second": 2.231,
"eval_weighted avg": {
"f1-score": 0.8613257791852104,
"precision": 0.8628576878135704,
"recall": 0.8602089552238806,
"support": 33500.0
},
"step": 1458
},
{
"epoch": 18.52,
"grad_norm": 1.7008241415023804,
"learning_rate": 1.2592592592592593e-05,
"loss": 0.0216,
"step": 1500
},
{
"epoch": 19.0,
"eval_Claim": {
"f1-score": 0.6404391582799633,
"precision": 0.6432509699816213,
"recall": 0.6376518218623481,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8286799195890104,
"precision": 0.8103975535168195,
"recall": 0.8478062157221207,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9006844397645144,
"precision": 0.9029750479846449,
"recall": 0.898405423469875,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8906363035888403,
"precision": 0.8907763936076507,
"recall": 0.8904962576262658,
"support": 15899.0
},
"eval_accuracy": 0.8528955223880597,
"eval_loss": 1.0161080360412598,
"eval_macro avg": {
"f1-score": 0.8151099553055821,
"precision": 0.8118499912726841,
"recall": 0.8185899296701524,
"support": 33500.0
},
"eval_runtime": 4.8924,
"eval_samples_per_second": 16.556,
"eval_steps_per_second": 2.248,
"eval_weighted avg": {
"f1-score": 0.8528362935634777,
"precision": 0.8528394387556787,
"recall": 0.8528955223880597,
"support": 33500.0
},
"step": 1539
},
{
"epoch": 20.0,
"eval_Claim": {
"f1-score": 0.650995688770273,
"precision": 0.6603498542274052,
"recall": 0.6419028340080971,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8586199145875477,
"precision": 0.8447589562140646,
"recall": 0.8729433272394881,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9104241552839684,
"precision": 0.913972286374134,
"recall": 0.9069034660555715,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8906836964688204,
"precision": 0.8866313493300093,
"recall": 0.894773256179634,
"support": 15899.0
},
"eval_accuracy": 0.8598507462686568,
"eval_loss": 0.9872655272483826,
"eval_macro avg": {
"f1-score": 0.8276808637776524,
"precision": 0.8264281115364033,
"recall": 0.8291307208706977,
"support": 33500.0
},
"eval_runtime": 4.9351,
"eval_samples_per_second": 16.413,
"eval_steps_per_second": 2.229,
"eval_weighted avg": {
"f1-score": 0.8594158610175964,
"precision": 0.8590759538589217,
"recall": 0.8598507462686568,
"support": 33500.0
},
"step": 1620
},
{
"epoch": 21.0,
"eval_Claim": {
"f1-score": 0.6439562566534404,
"precision": 0.6169108103096607,
"recall": 0.6734817813765183,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8524001794526693,
"precision": 0.8370044052863436,
"recall": 0.8683729433272395,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9078541717641301,
"precision": 0.9311383256374222,
"recall": 0.8857061014036093,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8851262038144394,
"precision": 0.8857952755905512,
"recall": 0.8844581420215107,
"support": 15899.0
},
"eval_accuracy": 0.8526865671641791,
"eval_loss": 1.0280390977859497,
"eval_macro avg": {
"f1-score": 0.8223342029211698,
"precision": 0.8177122042059943,
"recall": 0.8280047420322194,
"support": 33500.0
},
"eval_runtime": 4.9316,
"eval_samples_per_second": 16.425,
"eval_steps_per_second": 2.231,
"eval_weighted avg": {
"f1-score": 0.8545305598758789,
"precision": 0.8571336063495862,
"recall": 0.8526865671641791,
"support": 33500.0
},
"step": 1701
},
{
"epoch": 22.0,
"eval_Claim": {
"f1-score": 0.6535234899328859,
"precision": 0.6779808529155787,
"recall": 0.6307692307692307,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8511305126483099,
"precision": 0.8341377797279509,
"recall": 0.8688299817184644,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9092228079508632,
"precision": 0.9212073696589572,
"recall": 0.8975460708488494,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8968440594059406,
"precision": 0.882589367273613,
"recall": 0.9115667652053588,
"support": 15899.0
},
"eval_accuracy": 0.8629850746268657,
"eval_loss": 1.0620397329330444,
"eval_macro avg": {
"f1-score": 0.8276802174844999,
"precision": 0.828978842394025,
"recall": 0.8271780121354758,
"support": 33500.0
},
"eval_runtime": 4.8953,
"eval_samples_per_second": 16.546,
"eval_steps_per_second": 2.247,
"eval_weighted avg": {
"f1-score": 0.8618475752270865,
"precision": 0.8613257315871389,
"recall": 0.8629850746268657,
"support": 33500.0
},
"step": 1782
},
{
"epoch": 23.0,
"eval_Claim": {
"f1-score": 0.6436781609195402,
"precision": 0.5858518764530056,
"recall": 0.71417004048583,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8615044247787611,
"precision": 0.8349056603773585,
"recall": 0.8898537477148081,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9092565073582283,
"precision": 0.9129765113592607,
"recall": 0.905566695311754,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.873210033597547,
"precision": 0.9069657135113159,
"recall": 0.841876847600478,
"support": 15899.0
},
"eval_accuracy": 0.846089552238806,
"eval_loss": 1.189612627029419,
"eval_macro avg": {
"f1-score": 0.8219122816635192,
"precision": 0.8101749404252352,
"recall": 0.8378668327782175,
"support": 33500.0
},
"eval_runtime": 4.9275,
"eval_samples_per_second": 16.438,
"eval_steps_per_second": 2.232,
"eval_weighted avg": {
"f1-score": 0.8498672096145845,
"precision": 0.8567860518556853,
"recall": 0.846089552238806,
"support": 33500.0
},
"step": 1863
},
{
"epoch": 24.0,
"eval_Claim": {
"f1-score": 0.6632010396880935,
"precision": 0.6551451708473237,
"recall": 0.6714574898785425,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.84557309540151,
"precision": 0.8465414567109483,
"recall": 0.8446069469835467,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.911568570600174,
"precision": 0.922805987672439,
"recall": 0.9006015468347178,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8970938243768007,
"precision": 0.8933449759870268,
"recall": 0.9008742688219384,
"support": 15899.0
},
"eval_accuracy": 0.8632835820895522,
"eval_loss": 1.108552098274231,
"eval_macro avg": {
"f1-score": 0.8293591325166445,
"precision": 0.8294593978044343,
"recall": 0.8293850631296864,
"support": 33500.0
},
"eval_runtime": 4.9242,
"eval_samples_per_second": 16.449,
"eval_steps_per_second": 2.234,
"eval_weighted avg": {
"f1-score": 0.8637635648495541,
"precision": 0.864372798011359,
"recall": 0.8632835820895522,
"support": 33500.0
},
"step": 1944
},
{
"epoch": 24.69,
"grad_norm": 0.41564273834228516,
"learning_rate": 1.0123456790123458e-05,
"loss": 0.0102,
"step": 2000
},
{
"epoch": 25.0,
"eval_Claim": {
"f1-score": 0.6543478260869565,
"precision": 0.6391891891891892,
"recall": 0.6702429149797571,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8417964342134958,
"precision": 0.8314757021845742,
"recall": 0.8523765996343693,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9077508785630614,
"precision": 0.9285072391412881,
"recall": 0.8879022247684523,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8887081130127343,
"precision": 0.884198729921554,
"recall": 0.8932637272784452,
"support": 15899.0
},
"eval_accuracy": 0.8560298507462687,
"eval_loss": 1.0723340511322021,
"eval_macro avg": {
"f1-score": 0.8231508129690621,
"precision": 0.8208427151091513,
"recall": 0.8259463666652559,
"support": 33500.0
},
"eval_runtime": 4.8953,
"eval_samples_per_second": 16.546,
"eval_steps_per_second": 2.247,
"eval_weighted avg": {
"f1-score": 0.8570380328032566,
"precision": 0.8584774732395206,
"recall": 0.8560298507462687,
"support": 33500.0
},
"step": 2025
},
{
"epoch": 26.0,
"eval_Claim": {
"f1-score": 0.6502824298880191,
"precision": 0.636963696369637,
"recall": 0.66417004048583,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8458110516934045,
"precision": 0.8252173913043478,
"recall": 0.8674588665447898,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.912881895844394,
"precision": 0.9229943392543432,
"recall": 0.9029886374486775,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8878304207936407,
"precision": 0.8905271151047269,
"recall": 0.8851500094345557,
"support": 15899.0
},
"eval_accuracy": 0.8569850746268657,
"eval_loss": 1.126774787902832,
"eval_macro avg": {
"f1-score": 0.8242014495548646,
"precision": 0.8189256355082637,
"recall": 0.8299418884784633,
"support": 33500.0
},
"eval_runtime": 4.9541,
"eval_samples_per_second": 16.35,
"eval_steps_per_second": 2.22,
"eval_weighted avg": {
"f1-score": 0.8578882907501019,
"precision": 0.8590204964268868,
"recall": 0.8569850746268657,
"support": 33500.0
},
"step": 2106
},
{
"epoch": 27.0,
"eval_Claim": {
"f1-score": 0.6497180730042537,
"precision": 0.6353259818146644,
"recall": 0.6647773279352227,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8471862491170237,
"precision": 0.8737251092763477,
"recall": 0.8222120658135283,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9101418812401472,
"precision": 0.9107074569789675,
"recall": 0.9095770075432064,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.889659739522563,
"precision": 0.8921072603086264,
"recall": 0.8872256116736902,
"support": 15899.0
},
"eval_accuracy": 0.8571641791044776,
"eval_loss": 1.101758360862732,
"eval_macro avg": {
"f1-score": 0.824176485720997,
"precision": 0.8279664520946515,
"recall": 0.8209480032414119,
"support": 33500.0
},
"eval_runtime": 4.9026,
"eval_samples_per_second": 16.522,
"eval_steps_per_second": 2.244,
"eval_weighted avg": {
"f1-score": 0.8579064750628762,
"precision": 0.8588559229208557,
"recall": 0.8571641791044776,
"support": 33500.0
},
"step": 2187
},
{
"epoch": 28.0,
"eval_Claim": {
"f1-score": 0.6493561775034585,
"precision": 0.6845411711913844,
"recall": 0.6176113360323887,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.846973475402403,
"precision": 0.8403058929374719,
"recall": 0.8537477148080439,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9125746196803388,
"precision": 0.9202835226721041,
"recall": 0.9049937935644037,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8957433682911783,
"precision": 0.8788814236426367,
"recall": 0.9132649852191962,
"support": 15899.0
},
"eval_accuracy": 0.8631940298507462,
"eval_loss": 1.0291390419006348,
"eval_macro avg": {
"f1-score": 0.8261619102193447,
"precision": 0.8310030026108993,
"recall": 0.8224044574060082,
"support": 33500.0
},
"eval_runtime": 4.895,
"eval_samples_per_second": 16.548,
"eval_steps_per_second": 2.247,
"eval_weighted avg": {
"f1-score": 0.8614870234454082,
"precision": 0.8606473960260256,
"recall": 0.8631940298507462,
"support": 33500.0
},
"step": 2268
},
{
"epoch": 29.0,
"eval_Claim": {
"f1-score": 0.6583255135022417,
"precision": 0.6787787572565039,
"recall": 0.6390688259109312,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8515981735159819,
"precision": 0.8508211678832117,
"recall": 0.8523765996343693,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9123429557878584,
"precision": 0.916385704652731,
"recall": 0.9083357204239473,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8947940947940948,
"precision": 0.8844310641435242,
"recall": 0.9054028555255047,
"support": 15899.0
},
"eval_accuracy": 0.8635820895522388,
"eval_loss": 1.1136515140533447,
"eval_macro avg": {
"f1-score": 0.8292651844000442,
"precision": 0.8326041734839926,
"recall": 0.8262960003736881,
"support": 33500.0
},
"eval_runtime": 4.9179,
"eval_samples_per_second": 16.47,
"eval_steps_per_second": 2.237,
"eval_weighted avg": {
"f1-score": 0.8625887740134804,
"precision": 0.8618997238752699,
"recall": 0.8635820895522388,
"support": 33500.0
},
"step": 2349
},
{
"epoch": 30.0,
"eval_Claim": {
"f1-score": 0.6518860985604706,
"precision": 0.677736508630107,
"recall": 0.6279352226720648,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8529746979712788,
"precision": 0.8508412914961346,
"recall": 0.8551188299817185,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.911866692352148,
"precision": 0.919930035953746,
"recall": 0.9039434736942614,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8956451812445873,
"precision": 0.8810929227773383,
"recall": 0.9106862066796654,
"support": 15899.0
},
"eval_accuracy": 0.8632537313432835,
"eval_loss": 1.123184084892273,
"eval_macro avg": {
"f1-score": 0.8280931675321211,
"precision": 0.8324001897143315,
"recall": 0.8244209332569276,
"support": 33500.0
},
"eval_runtime": 4.9146,
"eval_samples_per_second": 16.482,
"eval_steps_per_second": 2.238,
"eval_weighted avg": {
"f1-score": 0.8619841364675112,
"precision": 0.8612711266927389,
"recall": 0.8632537313432835,
"support": 33500.0
},
"step": 2430
},
{
"epoch": 30.86,
"grad_norm": 0.36268848180770874,
"learning_rate": 7.654320987654322e-06,
"loss": 0.0075,
"step": 2500
},
{
"epoch": 31.0,
"eval_Claim": {
"f1-score": 0.6331647476595287,
"precision": 0.6758557316793017,
"recall": 0.5955465587044534,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8522257916475449,
"precision": 0.8557603686635945,
"recall": 0.8487202925045704,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9158597986057321,
"precision": 0.928901109692625,
"recall": 0.9031796046977943,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8934022573639618,
"precision": 0.8695962843872812,
"recall": 0.9185483363733569,
"support": 15899.0
},
"eval_accuracy": 0.8615522388059701,
"eval_loss": 1.120849370956421,
"eval_macro avg": {
"f1-score": 0.8236631488191918,
"precision": 0.8325283736057006,
"recall": 0.8164986980700437,
"support": 33500.0
},
"eval_runtime": 4.9277,
"eval_samples_per_second": 16.438,
"eval_steps_per_second": 2.232,
"eval_weighted avg": {
"f1-score": 0.8593583894385183,
"precision": 0.858663392490028,
"recall": 0.8615522388059701,
"support": 33500.0
},
"step": 2511
},
{
"epoch": 32.0,
"eval_Claim": {
"f1-score": 0.6540556600970584,
"precision": 0.6402947450067868,
"recall": 0.6684210526315789,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8491992882562277,
"precision": 0.8271230502599654,
"recall": 0.8724862888482633,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9160624787114983,
"precision": 0.934014685453463,
"recall": 0.8987873579681085,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8915745856353591,
"precision": 0.8899542520523908,
"recall": 0.8932008302408957,
"support": 15899.0
},
"eval_accuracy": 0.8604477611940299,
"eval_loss": 1.1111246347427368,
"eval_macro avg": {
"f1-score": 0.8277230031750359,
"precision": 0.8228466831931516,
"recall": 0.8332238824222116,
"support": 33500.0
},
"eval_runtime": 4.919,
"eval_samples_per_second": 16.467,
"eval_steps_per_second": 2.236,
"eval_weighted avg": {
"f1-score": 0.8614373038849609,
"precision": 0.8628095441324599,
"recall": 0.8604477611940299,
"support": 33500.0
},
"step": 2592
},
{
"epoch": 33.0,
"eval_Claim": {
"f1-score": 0.6564330438295295,
"precision": 0.6549778315195486,
"recall": 0.6578947368421053,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8466696468484577,
"precision": 0.8285214348206474,
"recall": 0.8656307129798904,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9148585819449826,
"precision": 0.9282555282555283,
"recall": 0.9018428339539769,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8948586439829872,
"precision": 0.8899048329912297,
"recall": 0.899867916221146,
"support": 15899.0
},
"eval_accuracy": 0.8625671641791045,
"eval_loss": 1.145164132118225,
"eval_macro avg": {
"f1-score": 0.8282049791514892,
"precision": 0.8254149068967386,
"recall": 0.8313090499992797,
"support": 33500.0
},
"eval_runtime": 5.0162,
"eval_samples_per_second": 16.148,
"eval_steps_per_second": 2.193,
"eval_weighted avg": {
"f1-score": 0.862804893528884,
"precision": 0.8632421634102047,
"recall": 0.8625671641791045,
"support": 33500.0
},
"step": 2673
},
{
"epoch": 34.0,
"eval_Claim": {
"f1-score": 0.6356152512998267,
"precision": 0.6835973904939422,
"recall": 0.5939271255060729,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8559207564160288,
"precision": 0.8433895297249334,
"recall": 0.8688299817184644,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9127349443805141,
"precision": 0.9166907444861794,
"recall": 0.9088131385467393,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8947336002463814,
"precision": 0.876591635990586,
"recall": 0.9136423674444933,
"support": 15899.0
},
"eval_accuracy": 0.8620597014925373,
"eval_loss": 1.1328134536743164,
"eval_macro avg": {
"f1-score": 0.8247511380856878,
"precision": 0.8300673251739102,
"recall": 0.8213031533039424,
"support": 33500.0
},
"eval_runtime": 4.886,
"eval_samples_per_second": 16.578,
"eval_steps_per_second": 2.251,
"eval_weighted avg": {
"f1-score": 0.8596160160977241,
"precision": 0.858499701125263,
"recall": 0.8620597014925373,
"support": 33500.0
},
"step": 2754
},
{
"epoch": 35.0,
"eval_Claim": {
"f1-score": 0.6474583247581807,
"precision": 0.6584344914190038,
"recall": 0.6368421052631579,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8435185185185186,
"precision": 0.8545966228893058,
"recall": 0.8327239488117002,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9133458015633241,
"precision": 0.9174373795761079,
"recall": 0.9092905566695312,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8930206484163319,
"precision": 0.8844540407156076,
"recall": 0.9017548273476319,
"support": 15899.0
},
"eval_accuracy": 0.8605373134328358,
"eval_loss": 1.16542649269104,
"eval_macro avg": {
"f1-score": 0.8243358233140887,
"precision": 0.8287306336500063,
"recall": 0.8201528595230052,
"support": 33500.0
},
"eval_runtime": 4.925,
"eval_samples_per_second": 16.447,
"eval_steps_per_second": 2.233,
"eval_weighted avg": {
"f1-score": 0.8599304033363548,
"precision": 0.8594859781531254,
"recall": 0.8605373134328358,
"support": 33500.0
},
"step": 2835
},
{
"epoch": 36.0,
"eval_Claim": {
"f1-score": 0.63220675944334,
"precision": 0.654320987654321,
"recall": 0.6115384615384616,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8513604677310547,
"precision": 0.8379814077025233,
"recall": 0.8651736745886655,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9143658430443401,
"precision": 0.9262343260188087,
"recall": 0.9027976701995608,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8918458920006188,
"precision": 0.8778021442495126,
"recall": 0.9063463110887477,
"support": 15899.0
},
"eval_accuracy": 0.8590746268656716,
"eval_loss": 1.1980067491531372,
"eval_macro avg": {
"f1-score": 0.8224447405548384,
"precision": 0.8240847164062914,
"recall": 0.8214640293538589,
"support": 33500.0
},
"eval_runtime": 4.9337,
"eval_samples_per_second": 16.418,
"eval_steps_per_second": 2.23,
"eval_weighted avg": {
"f1-score": 0.857954907646772,
"precision": 0.8573873846830882,
"recall": 0.8590746268656716,
"support": 33500.0
},
"step": 2916
},
{
"epoch": 37.0,
"eval_Claim": {
"f1-score": 0.6368211365066052,
"precision": 0.6605045672031318,
"recall": 0.6147773279352227,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8573378839590444,
"precision": 0.8536474852741278,
"recall": 0.8610603290676416,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9157544485701886,
"precision": 0.9250779423226813,
"recall": 0.9066170151818963,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8922981750695949,
"precision": 0.8778528391455176,
"recall": 0.9072268696144412,
"support": 15899.0
},
"eval_accuracy": 0.8608955223880597,
"eval_loss": 1.1901870965957642,
"eval_macro avg": {
"f1-score": 0.8255529110263582,
"precision": 0.8292707084863646,
"recall": 0.8224203854498005,
"support": 33500.0
},
"eval_runtime": 4.927,
"eval_samples_per_second": 16.44,
"eval_steps_per_second": 2.233,
"eval_weighted avg": {
"f1-score": 0.859674529246331,
"precision": 0.8589849802771131,
"recall": 0.8608955223880597,
"support": 33500.0
},
"step": 2997
},
{
"epoch": 37.04,
"grad_norm": 0.8262068033218384,
"learning_rate": 5.185185185185185e-06,
"loss": 0.0042,
"step": 3000
},
{
"epoch": 38.0,
"eval_Claim": {
"f1-score": 0.6409235668789809,
"precision": 0.6303837118245889,
"recall": 0.6518218623481782,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8523321956769055,
"precision": 0.8486633439057544,
"recall": 0.8560329067641682,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9147144240077445,
"precision": 0.9275547266123491,
"recall": 0.9022247684522104,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8904285669498698,
"precision": 0.8876734591823978,
"recall": 0.8932008302408957,
"support": 15899.0
},
"eval_accuracy": 0.858,
"eval_loss": 1.198099136352539,
"eval_macro avg": {
"f1-score": 0.8245996883783752,
"precision": 0.8235688103812725,
"recall": 0.8258200919513632,
"support": 33500.0
},
"eval_runtime": 4.9138,
"eval_samples_per_second": 16.484,
"eval_steps_per_second": 2.239,
"eval_weighted avg": {
"f1-score": 0.8587400959132037,
"precision": 0.8596528929024279,
"recall": 0.858,
"support": 33500.0
},
"step": 3078
},
{
"epoch": 39.0,
"eval_Claim": {
"f1-score": 0.6499377851513894,
"precision": 0.6662414965986394,
"recall": 0.6344129554655871,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8481362172112287,
"precision": 0.8540315106580166,
"recall": 0.8423217550274223,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9112606807007495,
"precision": 0.9111302023673158,
"recall": 0.9113911964098157,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8928604847010387,
"precision": 0.8855958420987502,
"recall": 0.9002452984464432,
"support": 15899.0
},
"eval_accuracy": 0.8607462686567164,
"eval_loss": 1.214090347290039,
"eval_macro avg": {
"f1-score": 0.8255487919411015,
"precision": 0.8292497629306805,
"recall": 0.822092801337317,
"support": 33500.0
},
"eval_runtime": 4.9164,
"eval_samples_per_second": 16.475,
"eval_steps_per_second": 2.237,
"eval_weighted avg": {
"f1-score": 0.8598697509596058,
"precision": 0.8591704131772522,
"recall": 0.8607462686567164,
"support": 33500.0
},
"step": 3159
},
{
"epoch": 40.0,
"eval_Claim": {
"f1-score": 0.6468762929251138,
"precision": 0.6613790186125211,
"recall": 0.632995951417004,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8509220173294822,
"precision": 0.8279290964115866,
"recall": 0.8752285191956124,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9132591629340654,
"precision": 0.921379980563654,
"recall": 0.9052802444380789,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8917924410627417,
"precision": 0.8843465891520812,
"recall": 0.8993647399207497,
"support": 15899.0
},
"eval_accuracy": 0.8603582089552239,
"eval_loss": 1.218752145767212,
"eval_macro avg": {
"f1-score": 0.8257124785628508,
"precision": 0.8237586711849608,
"recall": 0.8282173637428613,
"support": 33500.0
},
"eval_runtime": 5.1955,
"eval_samples_per_second": 15.59,
"eval_steps_per_second": 2.117,
"eval_weighted avg": {
"f1-score": 0.8597181341740886,
"precision": 0.8593600051422834,
"recall": 0.8603582089552239,
"support": 33500.0
},
"step": 3240
},
{
"epoch": 41.0,
"eval_Claim": {
"f1-score": 0.646806774127729,
"precision": 0.6519950637597697,
"recall": 0.6417004048582996,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8511878081577768,
"precision": 0.8350923482849604,
"recall": 0.8679159049360147,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9144551375522505,
"precision": 0.9312939312939313,
"recall": 0.8982144562207581,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8959641813320066,
"precision": 0.8859374039230155,
"recall": 0.9062205170136487,
"support": 15899.0
},
"eval_accuracy": 0.8622089552238806,
"eval_loss": 1.1967241764068604,
"eval_macro avg": {
"f1-score": 0.8271034752924407,
"precision": 0.8260796868154192,
"recall": 0.8285128207571802,
"support": 33500.0
},
"eval_runtime": 4.9208,
"eval_samples_per_second": 16.461,
"eval_steps_per_second": 2.235,
"eval_weighted avg": {
"f1-score": 0.8620790317320146,
"precision": 0.8622984418040036,
"recall": 0.8622089552238806,
"support": 33500.0
},
"step": 3321
},
{
"epoch": 42.0,
"eval_Claim": {
"f1-score": 0.6463245492371706,
"precision": 0.63290648040357,
"recall": 0.6603238866396761,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8430155210643016,
"precision": 0.8186907838070628,
"recall": 0.8688299817184644,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9103991443434294,
"precision": 0.9273969889064976,
"recall": 0.8940131767401891,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8930782040405945,
"precision": 0.8922651933701657,
"recall": 0.8938926976539405,
"support": 15899.0
},
"eval_accuracy": 0.8578507462686568,
"eval_loss": 1.2189435958862305,
"eval_macro avg": {
"f1-score": 0.8232043546713739,
"precision": 0.817814861621824,
"recall": 0.8292649356880676,
"support": 33500.0
},
"eval_runtime": 4.9128,
"eval_samples_per_second": 16.488,
"eval_steps_per_second": 2.239,
"eval_weighted avg": {
"f1-score": 0.8588364727782227,
"precision": 0.8601972066380149,
"recall": 0.8578507462686568,
"support": 33500.0
},
"step": 3402
},
{
"epoch": 43.0,
"eval_Claim": {
"f1-score": 0.6505707647237094,
"precision": 0.6493244605767292,
"recall": 0.6518218623481782,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8504863153132776,
"precision": 0.8419167039856695,
"recall": 0.8592321755027422,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9151807228915662,
"precision": 0.9239077551814732,
"recall": 0.9066170151818963,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8942687128092702,
"precision": 0.8905869877113093,
"recall": 0.8979810050946601,
"support": 15899.0
},
"eval_accuracy": 0.8618507462686568,
"eval_loss": 1.2199524641036987,
"eval_macro avg": {
"f1-score": 0.8276266289344558,
"precision": 0.8264339768637954,
"recall": 0.8289130145318692,
"support": 33500.0
},
"eval_runtime": 4.8976,
"eval_samples_per_second": 16.539,
"eval_steps_per_second": 2.246,
"eval_weighted avg": {
"f1-score": 0.8620104361623443,
"precision": 0.8622479110808169,
"recall": 0.8618507462686568,
"support": 33500.0
},
"step": 3483
},
{
"epoch": 43.21,
"grad_norm": 0.7243998050689697,
"learning_rate": 2.7160493827160496e-06,
"loss": 0.0038,
"step": 3500
},
{
"epoch": 44.0,
"eval_Claim": {
"f1-score": 0.649113924050633,
"precision": 0.6494427558257345,
"recall": 0.6487854251012146,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8521934758155231,
"precision": 0.8391670358883474,
"recall": 0.8656307129798904,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9142580488985719,
"precision": 0.9273227263798861,
"recall": 0.9015563830803017,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8929898516783762,
"precision": 0.886704700483691,
"recall": 0.8993647399207497,
"support": 15899.0
},
"eval_accuracy": 0.8608955223880597,
"eval_loss": 1.2192212343215942,
"eval_macro avg": {
"f1-score": 0.827138825110776,
"precision": 0.8256593046444147,
"recall": 0.8288343152705391,
"support": 33500.0
},
"eval_runtime": 4.9247,
"eval_samples_per_second": 16.448,
"eval_steps_per_second": 2.234,
"eval_weighted avg": {
"f1-score": 0.8610117106819026,
"precision": 0.8613108547662561,
"recall": 0.8608955223880597,
"support": 33500.0
},
"step": 3564
},
{
"epoch": 45.0,
"eval_Claim": {
"f1-score": 0.6460396039603961,
"precision": 0.6585365853658537,
"recall": 0.6340080971659919,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.85500340367597,
"precision": 0.8490310950878774,
"recall": 0.8610603290676416,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9148556835156363,
"precision": 0.923443579766537,
"recall": 0.9064260479327795,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8931060228969636,
"precision": 0.8835949522930132,
"recall": 0.902824076985974,
"support": 15899.0
},
"eval_accuracy": 0.8615820895522388,
"eval_loss": 1.2218410968780518,
"eval_macro avg": {
"f1-score": 0.8272511785122416,
"precision": 0.8286515531283204,
"recall": 0.8260796377880968,
"support": 33500.0
},
"eval_runtime": 4.9427,
"eval_samples_per_second": 16.388,
"eval_steps_per_second": 2.226,
"eval_weighted avg": {
"f1-score": 0.8609838603673272,
"precision": 0.8606075082137657,
"recall": 0.8615820895522388,
"support": 33500.0
},
"step": 3645
},
{
"epoch": 46.0,
"eval_Claim": {
"f1-score": 0.6521695155254374,
"precision": 0.6517081059227815,
"recall": 0.6526315789473685,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8555098308184728,
"precision": 0.8559011893870082,
"recall": 0.8551188299817185,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9134518133217743,
"precision": 0.9215743440233236,
"recall": 0.9054712116871957,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8919814861145858,
"precision": 0.8870436026621882,
"recall": 0.8969746524938675,
"support": 15899.0
},
"eval_accuracy": 0.8608656716417911,
"eval_loss": 1.2380608320236206,
"eval_macro avg": {
"f1-score": 0.8282781614450676,
"precision": 0.8290568104988254,
"recall": 0.8275490682775376,
"support": 33500.0
},
"eval_runtime": 4.9319,
"eval_samples_per_second": 16.424,
"eval_steps_per_second": 2.23,
"eval_weighted avg": {
"f1-score": 0.8609482807516783,
"precision": 0.8611016175916333,
"recall": 0.8608656716417911,
"support": 33500.0
},
"step": 3726
},
{
"epoch": 47.0,
"eval_Claim": {
"f1-score": 0.6483549652882584,
"precision": 0.6445289057811562,
"recall": 0.6522267206477733,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8573378839590444,
"precision": 0.8536474852741278,
"recall": 0.8610603290676416,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9157894736842105,
"precision": 0.9263456090651558,
"recall": 0.9054712116871957,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8921016397546626,
"precision": 0.887712524132777,
"recall": 0.8965343732310208,
"support": 15899.0
},
"eval_accuracy": 0.8609850746268657,
"eval_loss": 1.2338788509368896,
"eval_macro avg": {
"f1-score": 0.8283959906715441,
"precision": 0.8280586310633042,
"recall": 0.8288231586584078,
"support": 33500.0
},
"eval_runtime": 5.0503,
"eval_samples_per_second": 16.039,
"eval_steps_per_second": 2.178,
"eval_weighted avg": {
"f1-score": 0.8612930133427016,
"precision": 0.861704850067615,
"recall": 0.8609850746268657,
"support": 33500.0
},
"step": 3807
},
{
"epoch": 48.0,
"eval_Claim": {
"f1-score": 0.6475201140645687,
"precision": 0.6515679442508711,
"recall": 0.6435222672064778,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8567530695770805,
"precision": 0.8524886877828054,
"recall": 0.8610603290676416,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9163250542560887,
"precision": 0.9257454687195479,
"recall": 0.9070944333046882,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8935346979530704,
"precision": 0.8866183664623196,
"recall": 0.9005597836341909,
"support": 15899.0
},
"eval_accuracy": 0.8621194029850746,
"eval_loss": 1.2307249307632446,
"eval_macro avg": {
"f1-score": 0.8285332339627021,
"precision": 0.8291051168038859,
"recall": 0.8280592033032497,
"support": 33500.0
},
"eval_runtime": 4.9169,
"eval_samples_per_second": 16.474,
"eval_steps_per_second": 2.237,
"eval_weighted avg": {
"f1-score": 0.8619792697221943,
"precision": 0.8619602565896216,
"recall": 0.8621194029850746,
"support": 33500.0
},
"step": 3888
},
{
"epoch": 49.0,
"eval_Claim": {
"f1-score": 0.6471544715447154,
"precision": 0.649795918367347,
"recall": 0.6445344129554655,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8557801498977969,
"precision": 0.8505643340857788,
"recall": 0.8610603290676416,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9165701042873696,
"precision": 0.9270436566070905,
"recall": 0.9063305643082211,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8934311124980496,
"precision": 0.8865972996407779,
"recall": 0.9003710925215422,
"support": 15899.0
},
"eval_accuracy": 0.8619402985074627,
"eval_loss": 1.230076551437378,
"eval_macro avg": {
"f1-score": 0.8282339595569829,
"precision": 0.8285003021752486,
"recall": 0.8280740997132177,
"support": 33500.0
},
"eval_runtime": 5.0916,
"eval_samples_per_second": 15.908,
"eval_steps_per_second": 2.16,
"eval_weighted avg": {
"f1-score": 0.8618892542452355,
"precision": 0.8619691129059451,
"recall": 0.8619402985074627,
"support": 33500.0
},
"step": 3969
},
{
"epoch": 49.38,
"grad_norm": 0.007535469718277454,
"learning_rate": 2.469135802469136e-07,
"loss": 0.0027,
"step": 4000
},
{
"epoch": 50.0,
"eval_Claim": {
"f1-score": 0.6468680316237584,
"precision": 0.6477872513195291,
"recall": 0.6459514170040486,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8557801498977969,
"precision": 0.8505643340857788,
"recall": 0.8610603290676416,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9166023166023166,
"precision": 0.926710256660486,
"recall": 0.9067124988064547,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8933804004873324,
"precision": 0.8874751737835154,
"recall": 0.8993647399207497,
"support": 15899.0
},
"eval_accuracy": 0.8617910447761195,
"eval_loss": 1.2317436933517456,
"eval_macro avg": {
"f1-score": 0.828157724652801,
"precision": 0.8281342539623273,
"recall": 0.8282722461997236,
"support": 33500.0
},
"eval_runtime": 4.9035,
"eval_samples_per_second": 16.519,
"eval_steps_per_second": 2.243,
"eval_weighted avg": {
"f1-score": 0.8618330177111018,
"precision": 0.8619853161339563,
"recall": 0.8617910447761195,
"support": 33500.0
},
"step": 4050
}
],
"logging_steps": 500,
"max_steps": 4050,
"num_input_tokens_seen": 0,
"num_train_epochs": 50,
"save_steps": 500,
"total_flos": 7167109193370000.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}