longformer-simple / checkpoint-2997 /trainer_state.json
Theoreticallyhugo's picture
Training in progress, epoch 37, checkpoint
d0a93ee verified
raw
history blame
No virus
49.9 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 37.0,
"eval_steps": 500,
"global_step": 2997,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_Claim": {
"f1-score": 0.38183790485045715,
"precision": 0.6497890295358649,
"recall": 0.27035330261136714,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.6266258607498087,
"precision": 0.5535653937140926,
"recall": 0.7219039224327898,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.8545803971812939,
"precision": 0.8443037974683544,
"recall": 0.8651102464332037,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.878949075575738,
"precision": 0.8314310260186549,
"recall": 0.9322278794550709,
"support": 14534.0
},
"eval_accuracy": 0.7960859220535504,
"eval_loss": 0.5359914898872375,
"eval_macro avg": {
"f1-score": 0.6854983095893245,
"precision": 0.7197723116842417,
"recall": 0.697398837733108,
"support": 29841.0
},
"eval_runtime": 4.8191,
"eval_samples_per_second": 16.601,
"eval_steps_per_second": 2.075,
"eval_weighted avg": {
"f1-score": 0.7769240850962494,
"precision": 0.7862232339538372,
"recall": 0.7960859220535504,
"support": 29841.0
},
"step": 81
},
{
"epoch": 2.0,
"eval_Claim": {
"f1-score": 0.6161300742974051,
"precision": 0.6048625792811839,
"recall": 0.6278253236778583,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.7920215150156881,
"precision": 0.8057455540355677,
"recall": 0.7787571617452622,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.8922796352583586,
"precision": 0.9209436566695947,
"recall": 0.865346067680698,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8914289590611538,
"precision": 0.8790554552143957,
"recall": 0.9041557726709784,
"support": 14534.0
},
"eval_accuracy": 0.8413927147213566,
"eval_loss": 0.443877637386322,
"eval_macro avg": {
"f1-score": 0.7979650459081514,
"precision": 0.8026518113001855,
"recall": 0.7940210814436992,
"support": 29841.0
},
"eval_runtime": 4.8087,
"eval_samples_per_second": 16.637,
"eval_steps_per_second": 2.08,
"eval_weighted avg": {
"f1-score": 0.842071433389123,
"precision": 0.8435143116581857,
"recall": 0.8413927147213566,
"support": 29841.0
},
"step": 162
},
{
"epoch": 3.0,
"eval_Claim": {
"f1-score": 0.6042799305957202,
"precision": 0.6389432485322897,
"recall": 0.5731841123546193,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8349557522123894,
"precision": 0.8382940915148823,
"recall": 0.8316438959894227,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.8941106484235574,
"precision": 0.9022691799735862,
"recall": 0.8860983374602052,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8917763490086511,
"precision": 0.8729980887102089,
"recall": 0.9113802119168846,
"support": 14534.0
},
"eval_accuracy": 0.8464863778023525,
"eval_loss": 0.4555923044681549,
"eval_macro avg": {
"f1-score": 0.8062806700600795,
"precision": 0.8131261521827418,
"recall": 0.8005766394302829,
"support": 29841.0
},
"eval_runtime": 4.7811,
"eval_samples_per_second": 16.733,
"eval_steps_per_second": 2.092,
"eval_weighted avg": {
"f1-score": 0.8442159482345275,
"precision": 0.8429359878649869,
"recall": 0.8464863778023525,
"support": 29841.0
},
"step": 243
},
{
"epoch": 4.0,
"eval_Claim": {
"f1-score": 0.6278026905829596,
"precision": 0.5981717011128775,
"recall": 0.6605222734254992,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8182207014542344,
"precision": 0.7947652679684254,
"recall": 0.8431026884089907,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9069328626209752,
"precision": 0.9373427277302466,
"recall": 0.8784341469166372,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.888919552918449,
"precision": 0.8913795489137954,
"recall": 0.8864730975643319,
"support": 14534.0
},
"eval_accuracy": 0.8463858449783854,
"eval_loss": 0.4377317428588867,
"eval_macro avg": {
"f1-score": 0.8104689518941546,
"precision": 0.8054148114313363,
"recall": 0.8171330515788647,
"support": 29841.0
},
"eval_runtime": 4.8109,
"eval_samples_per_second": 16.629,
"eval_steps_per_second": 2.079,
"eval_weighted avg": {
"f1-score": 0.8487883791626096,
"precision": 0.8523207959781866,
"recall": 0.8463858449783854,
"support": 29841.0
},
"step": 324
},
{
"epoch": 5.0,
"eval_Claim": {
"f1-score": 0.6247990569070839,
"precision": 0.6105990783410138,
"recall": 0.6396752249286811,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.7998289136013687,
"precision": 0.7769007062733694,
"recall": 0.8241516086381666,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9028039585296891,
"precision": 0.9020600353148911,
"recall": 0.9035491097747907,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8885327014878567,
"precision": 0.9001058948111542,
"recall": 0.8772533370028898,
"support": 14534.0
},
"eval_accuracy": 0.8444086994403672,
"eval_loss": 0.5236712694168091,
"eval_macro avg": {
"f1-score": 0.8039911576314996,
"precision": 0.7974164286851071,
"recall": 0.8111573200861321,
"support": 29841.0
},
"eval_runtime": 4.8153,
"eval_samples_per_second": 16.614,
"eval_steps_per_second": 2.077,
"eval_weighted avg": {
"f1-score": 0.8455693764619782,
"precision": 0.8470828034323642,
"recall": 0.8444086994403672,
"support": 29841.0
},
"step": 405
},
{
"epoch": 6.0,
"eval_Claim": {
"f1-score": 0.6166529266281945,
"precision": 0.5813095006800078,
"recall": 0.6565723063418916,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8074107959743824,
"precision": 0.8392772230147408,
"recall": 0.7778757161745262,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.8913107165875667,
"precision": 0.8858607034707664,
"recall": 0.8968282042212004,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8815305371596762,
"precision": 0.8981792217065334,
"recall": 0.865487821659557,
"support": 14534.0
},
"eval_accuracy": 0.8358298984618477,
"eval_loss": 0.5537309050559998,
"eval_macro avg": {
"f1-score": 0.7992262440874549,
"precision": 0.8011566622180121,
"recall": 0.7991910120992938,
"support": 29841.0
},
"eval_runtime": 4.7996,
"eval_samples_per_second": 16.668,
"eval_steps_per_second": 2.084,
"eval_weighted avg": {
"f1-score": 0.838225042631562,
"precision": 0.8418105575563007,
"recall": 0.8358298984618477,
"support": 29841.0
},
"step": 486
},
{
"epoch": 6.17,
"grad_norm": 6.804128646850586,
"learning_rate": 1.7530864197530865e-05,
"loss": 0.3426,
"step": 500
},
{
"epoch": 7.0,
"eval_Claim": {
"f1-score": 0.6206536493682782,
"precision": 0.5920318725099601,
"recall": 0.652183454026772,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8364370652905541,
"precision": 0.8519195612431444,
"recall": 0.8215072719259586,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9008658773634918,
"precision": 0.9000706214689266,
"recall": 0.9016625397948355,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8854242963272995,
"precision": 0.8978566881233642,
"recall": 0.8733314985551122,
"support": 14534.0
},
"eval_accuracy": 0.8436714587312758,
"eval_loss": 0.5738794207572937,
"eval_macro avg": {
"f1-score": 0.8108452220874058,
"precision": 0.8104696858363488,
"recall": 0.8121711910756696,
"support": 29841.0
},
"eval_runtime": 4.8167,
"eval_samples_per_second": 16.609,
"eval_steps_per_second": 2.076,
"eval_weighted avg": {
"f1-score": 0.8456551258354699,
"precision": 0.8482906998207675,
"recall": 0.8436714587312758,
"support": 29841.0
},
"step": 567
},
{
"epoch": 8.0,
"eval_Claim": {
"f1-score": 0.6410105757931845,
"precision": 0.5787798408488064,
"recall": 0.7182356813693219,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8115942028985507,
"precision": 0.8804123711340206,
"recall": 0.75275451740855,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.8960937967338637,
"precision": 0.9094220495386109,
"recall": 0.8831505718665251,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.891816143497758,
"precision": 0.9084939329050679,
"recall": 0.8757396449704142,
"support": 14534.0
},
"eval_accuracy": 0.8444422103816897,
"eval_loss": 0.5891872644424438,
"eval_macro avg": {
"f1-score": 0.8101286797308392,
"precision": 0.8192770486066265,
"recall": 0.8074701039037029,
"support": 29841.0
},
"eval_runtime": 4.799,
"eval_samples_per_second": 16.67,
"eval_steps_per_second": 2.084,
"eval_weighted avg": {
"f1-score": 0.8486317402219317,
"precision": 0.8562720628608399,
"recall": 0.8444422103816897,
"support": 29841.0
},
"step": 648
},
{
"epoch": 9.0,
"eval_Claim": {
"f1-score": 0.6284997910572503,
"precision": 0.5998005982053839,
"recall": 0.6600833881939873,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.816712834718375,
"precision": 0.857489093553078,
"recall": 0.7796386073159982,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.8971996170416466,
"precision": 0.9108249301421455,
"recall": 0.8839759462327555,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8942406935938898,
"precision": 0.8943022295623452,
"recall": 0.8941791660932985,
"support": 14534.0
},
"eval_accuracy": 0.8468214872155759,
"eval_loss": 0.7009620070457458,
"eval_macro avg": {
"f1-score": 0.8091632341027905,
"precision": 0.8156042128657381,
"recall": 0.8044692769590099,
"support": 29841.0
},
"eval_runtime": 4.8005,
"eval_samples_per_second": 16.665,
"eval_steps_per_second": 2.083,
"eval_weighted avg": {
"f1-score": 0.848605581671113,
"precision": 0.8512257939173797,
"recall": 0.8468214872155759,
"support": 29841.0
},
"step": 729
},
{
"epoch": 10.0,
"eval_Claim": {
"f1-score": 0.6189745836985101,
"precision": 0.6180266900021877,
"recall": 0.619925389510643,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8208921766420085,
"precision": 0.8353102189781022,
"recall": 0.8069634200088145,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.8951902368987796,
"precision": 0.9085610200364299,
"recall": 0.8822072868765476,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8963474827245805,
"precision": 0.8870174493027017,
"recall": 0.905875877253337,
"support": 14534.0
},
"eval_accuracy": 0.8479608592205355,
"eval_loss": 0.7416248917579651,
"eval_macro avg": {
"f1-score": 0.8078511199909697,
"precision": 0.8122288445798553,
"recall": 0.8037429934123356,
"support": 29841.0
},
"eval_runtime": 4.7804,
"eval_samples_per_second": 16.735,
"eval_steps_per_second": 2.092,
"eval_weighted avg": {
"f1-score": 0.8479238041544329,
"precision": 0.8481312332795722,
"recall": 0.8479608592205355,
"support": 29841.0
},
"step": 810
},
{
"epoch": 11.0,
"eval_Claim": {
"f1-score": 0.6350413962855225,
"precision": 0.6477973065510158,
"recall": 0.6227781435154707,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8403837767117314,
"precision": 0.8316788951230039,
"recall": 0.8492728074041428,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9076665470637484,
"precision": 0.920883387938357,
"recall": 0.894823723617498,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.9016170675363501,
"precision": 0.8904844987250033,
"recall": 0.9130315123159488,
"support": 14534.0
},
"eval_accuracy": 0.8586843604436849,
"eval_loss": 0.753268837928772,
"eval_macro avg": {
"f1-score": 0.8211771968993381,
"precision": 0.822711022084345,
"recall": 0.8199765467132651,
"support": 29841.0
},
"eval_runtime": 4.7829,
"eval_samples_per_second": 16.726,
"eval_steps_per_second": 2.091,
"eval_weighted avg": {
"f1-score": 0.8579718131916827,
"precision": 0.8575920866110881,
"recall": 0.8586843604436849,
"support": 29841.0
},
"step": 891
},
{
"epoch": 12.0,
"eval_Claim": {
"f1-score": 0.6230640668523677,
"precision": 0.6328655500226347,
"recall": 0.6135615536537196,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.837270341207349,
"precision": 0.8310898827616153,
"recall": 0.8435434111943587,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.8982022214588343,
"precision": 0.8730105731775181,
"recall": 0.9248909326730338,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8888346297394397,
"precision": 0.9013795542978422,
"recall": 0.8766340993532407,
"support": 14534.0
},
"eval_accuracy": 0.8476592607486344,
"eval_loss": 0.8125314712524414,
"eval_macro avg": {
"f1-score": 0.8118428148144977,
"precision": 0.8095863900649026,
"recall": 0.8146574992185882,
"support": 29841.0
},
"eval_runtime": 4.7824,
"eval_samples_per_second": 16.728,
"eval_steps_per_second": 2.091,
"eval_weighted avg": {
"f1-score": 0.846990546753497,
"precision": 0.8469677413197487,
"recall": 0.8476592607486344,
"support": 29841.0
},
"step": 972
},
{
"epoch": 12.35,
"grad_norm": 15.335172653198242,
"learning_rate": 1.506172839506173e-05,
"loss": 0.061,
"step": 1000
},
{
"epoch": 13.0,
"eval_Claim": {
"f1-score": 0.6158673240699238,
"precision": 0.6292649416075109,
"recall": 0.6030283080974326,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8147180492053124,
"precision": 0.8050774526678142,
"recall": 0.8245923314235346,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9060346875742457,
"precision": 0.9128665469778575,
"recall": 0.8993043273198915,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.9008830850011933,
"precision": 0.8929368029739777,
"recall": 0.9089720655015825,
"support": 14534.0
},
"eval_accuracy": 0.8530880332428538,
"eval_loss": 0.8537381887435913,
"eval_macro avg": {
"f1-score": 0.8093757864626687,
"precision": 0.8100364360567901,
"recall": 0.8089742580856103,
"support": 29841.0
},
"eval_runtime": 4.7933,
"eval_samples_per_second": 16.69,
"eval_steps_per_second": 2.086,
"eval_weighted avg": {
"f1-score": 0.8522709558043637,
"precision": 0.8516552983596964,
"recall": 0.8530880332428538,
"support": 29841.0
},
"step": 1053
},
{
"epoch": 14.0,
"eval_Claim": {
"f1-score": 0.6346033728919426,
"precision": 0.6036838978015449,
"recall": 0.6688610928242265,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8067737733391228,
"precision": 0.7950363714163458,
"recall": 0.8188629352137505,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9004597838418822,
"precision": 0.9121703363174449,
"recall": 0.8890461030538852,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8921770010096438,
"precision": 0.903023468884347,
"recall": 0.8815880005504335,
"support": 14534.0
},
"eval_accuracy": 0.8464528668610302,
"eval_loss": 0.9449928998947144,
"eval_macro avg": {
"f1-score": 0.8085034827706478,
"precision": 0.8034785186049206,
"recall": 0.8145895329105739,
"support": 29841.0
},
"eval_runtime": 4.8154,
"eval_samples_per_second": 16.613,
"eval_steps_per_second": 2.077,
"eval_weighted avg": {
"f1-score": 0.8487033685671466,
"precision": 0.8517001698367573,
"recall": 0.8464528668610302,
"support": 29841.0
},
"step": 1134
},
{
"epoch": 15.0,
"eval_Claim": {
"f1-score": 0.6062717770034843,
"precision": 0.5687367813881946,
"recall": 0.6491112574061882,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8176843948347561,
"precision": 0.8121739130434783,
"recall": 0.8232701630674306,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.8942823803967327,
"precision": 0.8850906571197598,
"recall": 0.9036670203985379,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.877618288144604,
"precision": 0.9049777063080184,
"recall": 0.8518645933672767,
"support": 14534.0
},
"eval_accuracy": 0.8334506216279616,
"eval_loss": 1.0531803369522095,
"eval_macro avg": {
"f1-score": 0.7989642100948944,
"precision": 0.7927447644648628,
"recall": 0.8069782585598584,
"support": 29841.0
},
"eval_runtime": 4.774,
"eval_samples_per_second": 16.757,
"eval_steps_per_second": 2.095,
"eval_weighted avg": {
"f1-score": 0.8363600230462553,
"precision": 0.8409220859889104,
"recall": 0.8334506216279616,
"support": 29841.0
},
"step": 1215
},
{
"epoch": 16.0,
"eval_Claim": {
"f1-score": 0.6349479110729246,
"precision": 0.6217921750105174,
"recall": 0.6486723721746763,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8260292164674634,
"precision": 0.8297020898176968,
"recall": 0.8223887174966946,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9045516104339634,
"precision": 0.9034344860032933,
"recall": 0.9056715010022403,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8956702459300312,
"precision": 0.90185546875,
"recall": 0.8895692858125774,
"support": 14534.0
},
"eval_accuracy": 0.8522502597097953,
"eval_loss": 0.9694080352783203,
"eval_macro avg": {
"f1-score": 0.8152997459760956,
"precision": 0.8141960548953769,
"recall": 0.8165754691215471,
"support": 29841.0
},
"eval_runtime": 4.7829,
"eval_samples_per_second": 16.726,
"eval_steps_per_second": 2.091,
"eval_weighted avg": {
"f1-score": 0.8530843968151707,
"precision": 0.854049671322198,
"recall": 0.8522502597097953,
"support": 29841.0
},
"step": 1296
},
{
"epoch": 17.0,
"eval_Claim": {
"f1-score": 0.6241452751933639,
"precision": 0.6379468377635197,
"recall": 0.6109282422646478,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8151592652871426,
"precision": 0.8626968503937008,
"recall": 0.7725870427501101,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9064921025764218,
"precision": 0.9237454100367197,
"recall": 0.8898714774201155,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8993928008319636,
"precision": 0.8775777414075286,
"recall": 0.9223200770606853,
"support": 14534.0
},
"eval_accuracy": 0.8541603833651688,
"eval_loss": 0.985948920249939,
"eval_macro avg": {
"f1-score": 0.8112973609722229,
"precision": 0.8254917099003672,
"recall": 0.7989267098738898,
"support": 29841.0
},
"eval_runtime": 4.7769,
"eval_samples_per_second": 16.747,
"eval_steps_per_second": 2.093,
"eval_weighted avg": {
"f1-score": 0.8529727851357218,
"precision": 0.85297351326598,
"recall": 0.8541603833651688,
"support": 29841.0
},
"step": 1377
},
{
"epoch": 18.0,
"eval_Claim": {
"f1-score": 0.638718890248878,
"precision": 0.5967219363445778,
"recall": 0.6870748299319728,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.778357854769815,
"precision": 0.8431876606683805,
"recall": 0.7227853680035258,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9013165698019214,
"precision": 0.9066937119675457,
"recall": 0.8960028298549699,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8934796194708702,
"precision": 0.9018082422203533,
"recall": 0.885303426448328,
"support": 14534.0
},
"eval_accuracy": 0.8457156261519386,
"eval_loss": 0.9986578822135925,
"eval_macro avg": {
"f1-score": 0.8029682335728712,
"precision": 0.8121028878002143,
"recall": 0.7977916135596992,
"support": 29841.0
},
"eval_runtime": 4.7848,
"eval_samples_per_second": 16.719,
"eval_steps_per_second": 2.09,
"eval_weighted avg": {
"f1-score": 0.8480491462490054,
"precision": 0.8521498954326654,
"recall": 0.8457156261519386,
"support": 29841.0
},
"step": 1458
},
{
"epoch": 18.52,
"grad_norm": 11.437110900878906,
"learning_rate": 1.2592592592592593e-05,
"loss": 0.0189,
"step": 1500
},
{
"epoch": 19.0,
"eval_Claim": {
"f1-score": 0.6127094530509011,
"precision": 0.5894160583941606,
"recall": 0.6379196840026333,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.800753453327752,
"precision": 0.762455161418892,
"recall": 0.8431026884089907,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.8959336706699994,
"precision": 0.9275435496086847,
"recall": 0.8664072632944229,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8912863642630635,
"precision": 0.8930100842657825,
"recall": 0.8895692858125774,
"support": 14534.0
},
"eval_accuracy": 0.8410240943668108,
"eval_loss": 1.0641764402389526,
"eval_macro avg": {
"f1-score": 0.800170735327929,
"precision": 0.79310621342188,
"recall": 0.8092497303796561,
"support": 29841.0
},
"eval_runtime": 4.8048,
"eval_samples_per_second": 16.65,
"eval_steps_per_second": 2.081,
"eval_weighted avg": {
"f1-score": 0.8431820663618932,
"precision": 0.8465361465202841,
"recall": 0.8410240943668108,
"support": 29841.0
},
"step": 1539
},
{
"epoch": 20.0,
"eval_Claim": {
"f1-score": 0.6346276257161044,
"precision": 0.6142945163277881,
"recall": 0.6563528637261357,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8171902268760908,
"precision": 0.8090712742980561,
"recall": 0.8254737769942706,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9044396961063628,
"precision": 0.9106011712680769,
"recall": 0.8983610423299139,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.894879267277269,
"precision": 0.902519244226732,
"recall": 0.8873675519471584,
"support": 14534.0
},
"eval_accuracy": 0.8505076907610335,
"eval_loss": 1.087249994277954,
"eval_macro avg": {
"f1-score": 0.8127842039939568,
"precision": 0.8091215515301633,
"recall": 0.8168888087493696,
"support": 29841.0
},
"eval_runtime": 4.7799,
"eval_samples_per_second": 16.737,
"eval_steps_per_second": 2.092,
"eval_weighted avg": {
"f1-score": 0.85194635060675,
"precision": 0.8536961248417916,
"recall": 0.8505076907610335,
"support": 29841.0
},
"step": 1620
},
{
"epoch": 21.0,
"eval_Claim": {
"f1-score": 0.6397356640375187,
"precision": 0.6219689119170985,
"recall": 0.6585472898836954,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8133363574506692,
"precision": 0.8378504672897197,
"recall": 0.7902159541648304,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9003324267150197,
"precision": 0.9236111111111112,
"recall": 0.8781983256691428,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8964765215020788,
"precision": 0.8880637321091007,
"recall": 0.9050502270538049,
"support": 14534.0
},
"eval_accuracy": 0.8510438658221909,
"eval_loss": 1.093902826309204,
"eval_macro avg": {
"f1-score": 0.8124702424263215,
"precision": 0.8178735556067575,
"recall": 0.8080029491928683,
"support": 29841.0
},
"eval_runtime": 4.813,
"eval_samples_per_second": 16.622,
"eval_steps_per_second": 2.078,
"eval_weighted avg": {
"f1-score": 0.8520439894961911,
"precision": 0.8537133191814482,
"recall": 0.8510438658221909,
"support": 29841.0
},
"step": 1701
},
{
"epoch": 22.0,
"eval_Claim": {
"f1-score": 0.6279694722132645,
"precision": 0.6154656552886641,
"recall": 0.640991880623217,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8173439463762043,
"precision": 0.7788423153692615,
"recall": 0.8598501542529748,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9038883941822499,
"precision": 0.910210425633668,
"recall": 0.8976535785874308,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8954102920723226,
"precision": 0.9051033319274567,
"recall": 0.8859226640979772,
"support": 14534.0
},
"eval_accuracy": 0.849870982875909,
"eval_loss": 1.0930607318878174,
"eval_macro avg": {
"f1-score": 0.8111530262110103,
"precision": 0.8024054320547626,
"recall": 0.8211045693904,
"support": 29841.0
},
"eval_runtime": 4.8038,
"eval_samples_per_second": 16.654,
"eval_steps_per_second": 2.082,
"eval_weighted avg": {
"f1-score": 0.8510432276144315,
"precision": 0.8527239921837771,
"recall": 0.849870982875909,
"support": 29841.0
},
"step": 1782
},
{
"epoch": 23.0,
"eval_Claim": {
"f1-score": 0.6225736879942487,
"precision": 0.5851351351351352,
"recall": 0.6651305683563749,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8058810995099083,
"precision": 0.7801155115511551,
"recall": 0.8334067871308947,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.8976647945610404,
"precision": 0.9001659947830211,
"recall": 0.8951774554887395,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8881674136288245,
"precision": 0.9116858653915816,
"recall": 0.8658318425760286,
"support": 14534.0
},
"eval_accuracy": 0.8410576053081331,
"eval_loss": 1.1820006370544434,
"eval_macro avg": {
"f1-score": 0.8035717489235055,
"precision": 0.7942756267152233,
"recall": 0.8148866633880094,
"support": 29841.0
},
"eval_runtime": 4.7901,
"eval_samples_per_second": 16.701,
"eval_steps_per_second": 2.088,
"eval_weighted avg": {
"f1-score": 0.8440512323089471,
"precision": 0.8485403664715132,
"recall": 0.8410576053081331,
"support": 29841.0
},
"step": 1863
},
{
"epoch": 24.0,
"eval_Claim": {
"f1-score": 0.6244661044792466,
"precision": 0.6233056405771753,
"recall": 0.6256308975202984,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8177083333333335,
"precision": 0.8054724241128688,
"recall": 0.8303217276333187,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9007499414108273,
"precision": 0.8951904040992198,
"recall": 0.9063789647447235,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8948225108225109,
"precision": 0.9008437347465309,
"recall": 0.888881243979634,
"support": 14534.0
},
"eval_accuracy": 0.8492007640494621,
"eval_loss": 1.1699731349945068,
"eval_macro avg": {
"f1-score": 0.8094367225114796,
"precision": 0.8062030508839487,
"recall": 0.8128032084694936,
"support": 29841.0
},
"eval_runtime": 4.802,
"eval_samples_per_second": 16.66,
"eval_steps_per_second": 2.082,
"eval_weighted avg": {
"f1-score": 0.8493576914930753,
"precision": 0.8496026739182283,
"recall": 0.8492007640494621,
"support": 29841.0
},
"step": 1944
},
{
"epoch": 24.69,
"grad_norm": 9.07358169555664,
"learning_rate": 1.0123456790123458e-05,
"loss": 0.006,
"step": 2000
},
{
"epoch": 25.0,
"eval_Claim": {
"f1-score": 0.622420524261015,
"precision": 0.632940108892922,
"recall": 0.6122448979591837,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8149307107733571,
"precision": 0.8267573696145125,
"recall": 0.8034376377258704,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9018776855612456,
"precision": 0.9004466384579219,
"recall": 0.9033132885272963,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8975866548164354,
"precision": 0.8919157608695653,
"recall": 0.9033301224714463,
"support": 14534.0
},
"eval_accuracy": 0.8512784424114473,
"eval_loss": 1.1391637325286865,
"eval_macro avg": {
"f1-score": 0.8092038938530133,
"precision": 0.8130149694587304,
"recall": 0.8055814866709491,
"support": 29841.0
},
"eval_runtime": 4.83,
"eval_samples_per_second": 16.563,
"eval_steps_per_second": 2.07,
"eval_weighted avg": {
"f1-score": 0.8505008948811765,
"precision": 0.8498378793311306,
"recall": 0.8512784424114473,
"support": 29841.0
},
"step": 2025
},
{
"epoch": 26.0,
"eval_Claim": {
"f1-score": 0.6216186854970125,
"precision": 0.6155335628227194,
"recall": 0.6278253236778583,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8021786492374727,
"precision": 0.7931925894011203,
"recall": 0.8113706478624945,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9000058854687776,
"precision": 0.8984723854289072,
"recall": 0.9015446291710884,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8959025470653377,
"precision": 0.9012672329759086,
"recall": 0.8906013485619926,
"support": 14534.0
},
"eval_accuracy": 0.8475587279246674,
"eval_loss": 1.18631911277771,
"eval_macro avg": {
"f1-score": 0.80492644181715,
"precision": 0.8021164426571639,
"recall": 0.8078354873183584,
"support": 29841.0
},
"eval_runtime": 4.7928,
"eval_samples_per_second": 16.692,
"eval_steps_per_second": 2.086,
"eval_weighted avg": {
"f1-score": 0.8480566079768785,
"precision": 0.848621115111045,
"recall": 0.8475587279246674,
"support": 29841.0
},
"step": 2106
},
{
"epoch": 27.0,
"eval_Claim": {
"f1-score": 0.6274287012839449,
"precision": 0.6505655042412818,
"recall": 0.6058810621022602,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8276779773785763,
"precision": 0.8330357142857143,
"recall": 0.8223887174966946,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9055736912102527,
"precision": 0.9291651159905719,
"recall": 0.8831505718665251,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8996312437143815,
"precision": 0.8772228033472803,
"recall": 0.9232145314435117,
"support": 14534.0
},
"eval_accuracy": 0.8557018866659964,
"eval_loss": 1.173492431640625,
"eval_macro avg": {
"f1-score": 0.8150779033967888,
"precision": 0.8224972844662122,
"recall": 0.8086587207272479,
"support": 29841.0
},
"eval_runtime": 4.7958,
"eval_samples_per_second": 16.681,
"eval_steps_per_second": 2.085,
"eval_weighted avg": {
"f1-score": 0.8542811867471567,
"precision": 0.8540126205927153,
"recall": 0.8557018866659964,
"support": 29841.0
},
"step": 2187
},
{
"epoch": 28.0,
"eval_Claim": {
"f1-score": 0.6397700895324416,
"precision": 0.644543429844098,
"recall": 0.6350669299978056,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8197434224831485,
"precision": 0.8090128755364807,
"recall": 0.8307624504186867,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9052350945225399,
"precision": 0.9310731646516266,
"recall": 0.8807923593915812,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.9018691588785046,
"precision": 0.8879183891185491,
"recall": 0.916265308930783,
"support": 14534.0
},
"eval_accuracy": 0.856740725846989,
"eval_loss": 1.153747797012329,
"eval_macro avg": {
"f1-score": 0.8166544413541587,
"precision": 0.8181369647876886,
"recall": 0.815721762184714,
"support": 29841.0
},
"eval_runtime": 4.8135,
"eval_samples_per_second": 16.62,
"eval_steps_per_second": 2.077,
"eval_weighted avg": {
"f1-score": 0.8565562586843419,
"precision": 0.8570179283955386,
"recall": 0.856740725846989,
"support": 29841.0
},
"step": 2268
},
{
"epoch": 29.0,
"eval_Claim": {
"f1-score": 0.6479684334008745,
"precision": 0.6302904564315353,
"recall": 0.6666666666666666,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8285593934288121,
"precision": 0.7934651068979427,
"recall": 0.8669017188188629,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9062767649577879,
"precision": 0.9417673235855054,
"recall": 0.8733639900955076,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8984971414877958,
"precision": 0.8941200517816993,
"recall": 0.9029172973716802,
"support": 14534.0
},
"eval_accuracy": 0.8557018866659964,
"eval_loss": 1.141819953918457,
"eval_macro avg": {
"f1-score": 0.8203254333188175,
"precision": 0.8149107346741707,
"recall": 0.8274624182381793,
"support": 29841.0
},
"eval_runtime": 4.7777,
"eval_samples_per_second": 16.744,
"eval_steps_per_second": 2.093,
"eval_weighted avg": {
"f1-score": 0.857132271461693,
"precision": 0.8597190255498753,
"recall": 0.8557018866659964,
"support": 29841.0
},
"step": 2349
},
{
"epoch": 30.0,
"eval_Claim": {
"f1-score": 0.6132195178221248,
"precision": 0.6464007782101168,
"recall": 0.5832784726793944,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8116131848941279,
"precision": 0.8040657439446367,
"recall": 0.8193036579991185,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9048840121653051,
"precision": 0.9154198841698842,
"recall": 0.8945879023700035,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.9030779085055456,
"precision": 0.8853195849031661,
"recall": 0.9215632310444475,
"support": 14534.0
},
"eval_accuracy": 0.8544619818370698,
"eval_loss": 1.1914094686508179,
"eval_macro avg": {
"f1-score": 0.8081986558467759,
"precision": 0.8128014978069509,
"recall": 0.804683316023241,
"support": 29841.0
},
"eval_runtime": 4.773,
"eval_samples_per_second": 16.761,
"eval_steps_per_second": 2.095,
"eval_weighted avg": {
"f1-score": 0.8523724837851865,
"precision": 0.8512108978901942,
"recall": 0.8544619818370698,
"support": 29841.0
},
"step": 2430
},
{
"epoch": 30.86,
"grad_norm": 0.007714413572102785,
"learning_rate": 7.654320987654322e-06,
"loss": 0.0028,
"step": 2500
},
{
"epoch": 31.0,
"eval_Claim": {
"f1-score": 0.6275360746446783,
"precision": 0.6206008583690987,
"recall": 0.6346280447662936,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8174338447854126,
"precision": 0.8249551166965888,
"recall": 0.8100484795063905,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9020403293163107,
"precision": 0.9129332206255283,
"recall": 0.8914043155288291,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8989933575292748,
"precision": 0.8947655398037078,
"recall": 0.9032613182881519,
"support": 14534.0
},
"eval_accuracy": 0.8517811065312825,
"eval_loss": 1.231397271156311,
"eval_macro avg": {
"f1-score": 0.8115009015689192,
"precision": 0.8133136838737309,
"recall": 0.8098355395224164,
"support": 29841.0
},
"eval_runtime": 4.7995,
"eval_samples_per_second": 16.668,
"eval_steps_per_second": 2.084,
"eval_weighted avg": {
"f1-score": 0.8522037725691503,
"precision": 0.8527532680206674,
"recall": 0.8517811065312825,
"support": 29841.0
},
"step": 2511
},
{
"epoch": 32.0,
"eval_Claim": {
"f1-score": 0.6187466523835029,
"precision": 0.6044370029300963,
"recall": 0.6337502743032697,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8172382671480145,
"precision": 0.8372630605640314,
"recall": 0.7981489643014544,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9010989010989012,
"precision": 0.9079482882451521,
"recall": 0.8943520811225092,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8937414030261347,
"precision": 0.8933727485219304,
"recall": 0.8941103619100041,
"support": 14534.0
},
"eval_accuracy": 0.847123085687477,
"eval_loss": 1.2467483282089233,
"eval_macro avg": {
"f1-score": 0.8077063059141383,
"precision": 0.8107552750653025,
"recall": 0.8050904204093093,
"support": 29841.0
},
"eval_runtime": 4.804,
"eval_samples_per_second": 16.653,
"eval_steps_per_second": 2.082,
"eval_weighted avg": {
"f1-score": 0.8480211606471663,
"precision": 0.8491256414462351,
"recall": 0.847123085687477,
"support": 29841.0
},
"step": 2592
},
{
"epoch": 33.0,
"eval_Claim": {
"f1-score": 0.6165579750193435,
"precision": 0.6211581291759466,
"recall": 0.6120254553434277,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8063751884557399,
"precision": 0.7885425442291492,
"recall": 0.8250330542089026,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.905199879771566,
"precision": 0.9233505028207015,
"recall": 0.8877490861926659,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.897230643458119,
"precision": 0.8884841125278283,
"recall": 0.9061510939865144,
"support": 14534.0
},
"eval_accuracy": 0.8498374719345867,
"eval_loss": 1.2608078718185425,
"eval_macro avg": {
"f1-score": 0.8063409216761921,
"precision": 0.8053838221884064,
"recall": 0.8077396724328777,
"support": 29841.0
},
"eval_runtime": 4.7947,
"eval_samples_per_second": 16.685,
"eval_steps_per_second": 2.086,
"eval_weighted avg": {
"f1-score": 0.8497258921327092,
"precision": 0.8499709906977834,
"recall": 0.8498374719345867,
"support": 29841.0
},
"step": 2673
},
{
"epoch": 34.0,
"eval_Claim": {
"f1-score": 0.6242761692650335,
"precision": 0.633732760569749,
"recall": 0.6150976519640114,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8270332187857962,
"precision": 0.8611641221374046,
"recall": 0.7955046275892463,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9015324995596267,
"precision": 0.8978947368421053,
"recall": 0.9051998585072515,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8962669760458609,
"precision": 0.8890468453831574,
"recall": 0.9036053392046236,
"support": 14534.0
},
"eval_accuracy": 0.8517811065312825,
"eval_loss": 1.2305262088775635,
"eval_macro avg": {
"f1-score": 0.8122772159140793,
"precision": 0.820459616233104,
"recall": 0.8048518693162833,
"support": 29841.0
},
"eval_runtime": 4.7809,
"eval_samples_per_second": 16.733,
"eval_steps_per_second": 2.092,
"eval_weighted avg": {
"f1-score": 0.8509636485165197,
"precision": 0.8504525215978962,
"recall": 0.8517811065312825,
"support": 29841.0
},
"step": 2754
},
{
"epoch": 35.0,
"eval_Claim": {
"f1-score": 0.633076593923834,
"precision": 0.6176163640158631,
"recall": 0.6493307000219443,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8311867525298988,
"precision": 0.8691678691678691,
"recall": 0.7963860731599823,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9039272770483037,
"precision": 0.9110179640718563,
"recall": 0.8969461148449476,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8963128108386212,
"precision": 0.8936461254360167,
"recall": 0.8989954589239025,
"support": 14534.0
},
"eval_accuracy": 0.8524848362990516,
"eval_loss": 1.2047479152679443,
"eval_macro avg": {
"f1-score": 0.8161258585851644,
"precision": 0.8228620806729012,
"recall": 0.8104145867376942,
"support": 29841.0
},
"eval_runtime": 4.7907,
"eval_samples_per_second": 16.699,
"eval_steps_per_second": 2.087,
"eval_weighted avg": {
"f1-score": 0.8533263097542454,
"precision": 0.8545697465347227,
"recall": 0.8524848362990516,
"support": 29841.0
},
"step": 2835
},
{
"epoch": 36.0,
"eval_Claim": {
"f1-score": 0.6346997752327946,
"precision": 0.6195152528207272,
"recall": 0.6506473557164801,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.8192934782608694,
"precision": 0.8425710293432697,
"recall": 0.7972675187307183,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9063178271836687,
"precision": 0.9018211533971515,
"recall": 0.9108595684471171,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.8959689707715751,
"precision": 0.9019662529633246,
"recall": 0.8900509150956378,
"support": 14534.0
},
"eval_accuracy": 0.8523507925337622,
"eval_loss": 1.2302902936935425,
"eval_macro avg": {
"f1-score": 0.8140700128622269,
"precision": 0.8164684221311183,
"recall": 0.8122063394974884,
"support": 29841.0
},
"eval_runtime": 4.803,
"eval_samples_per_second": 16.656,
"eval_steps_per_second": 2.082,
"eval_weighted avg": {
"f1-score": 0.8531818066233882,
"precision": 0.8542759088239045,
"recall": 0.8523507925337622,
"support": 29841.0
},
"step": 2916
},
{
"epoch": 37.0,
"eval_Claim": {
"f1-score": 0.6348107625320978,
"precision": 0.6461363636363636,
"recall": 0.6238753565942506,
"support": 4557.0
},
"eval_MajorClaim": {
"f1-score": 0.821917808219178,
"precision": 0.8526764566556134,
"recall": 0.7933010136624064,
"support": 2269.0
},
"eval_O": {
"f1-score": 0.9034164794893014,
"precision": 0.9057721939077872,
"recall": 0.9010729866760995,
"support": 8481.0
},
"eval_Premise": {
"f1-score": 0.9000577700751011,
"precision": 0.8892096958302558,
"recall": 0.9111737993670015,
"support": 14534.0
},
"eval_accuracy": 0.85546731007674,
"eval_loss": 1.2363109588623047,
"eval_macro avg": {
"f1-score": 0.8150507050789195,
"precision": 0.823448677507505,
"recall": 0.8073557890749394,
"support": 29841.0
},
"eval_runtime": 4.7822,
"eval_samples_per_second": 16.729,
"eval_steps_per_second": 2.091,
"eval_weighted avg": {
"f1-score": 0.8545651601664946,
"precision": 0.8540194358423435,
"recall": 0.85546731007674,
"support": 29841.0
},
"step": 2997
}
],
"logging_steps": 500,
"max_steps": 4050,
"num_input_tokens_seen": 0,
"num_train_epochs": 50,
"save_steps": 500,
"total_flos": 5320183110891600.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}