longformer-simple / checkpoint-1377 /trainer_state.json
Theoreticallyhugo's picture
Training in progress, epoch 17, checkpoint
9d65dda verified
raw
history blame
No virus
23.1 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 17.0,
"eval_steps": 500,
"global_step": 1377,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_Claim": {
"f1-score": 0.41092129347162903,
"precision": 0.5640703517587939,
"recall": 0.32317658349328215,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.6523102655237298,
"precision": 0.5929304446978335,
"recall": 0.724907063197026,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.8834432924913603,
"precision": 0.8545380875202593,
"recall": 0.9143724257533059,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8664166767469054,
"precision": 0.8441231929604023,
"recall": 0.8899196554294707,
"support": 12073.0
},
"eval_accuracy": 0.7997031029363844,
"eval_loss": 0.5104541182518005,
"eval_macro avg": {
"f1-score": 0.7032728820584061,
"precision": 0.7139155192343223,
"recall": 0.7130939319682712,
"support": 27619.0
},
"eval_runtime": 4.7759,
"eval_samples_per_second": 16.751,
"eval_steps_per_second": 2.094,
"eval_weighted avg": {
"f1-score": 0.7866826459135919,
"precision": 0.7857670171690952,
"recall": 0.7997031029363844,
"support": 27619.0
},
"step": 81
},
{
"epoch": 2.0,
"eval_Claim": {
"f1-score": 0.5200312174817897,
"precision": 0.5678977272727272,
"recall": 0.47960652591170827,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.7526302911671151,
"precision": 0.7948320413436692,
"recall": 0.7146840148698885,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9116096466629278,
"precision": 0.944560669456067,
"recall": 0.8808801213960546,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8738735224125151,
"precision": 0.8259587020648967,
"recall": 0.9276898865236478,
"support": 12073.0
},
"eval_accuracy": 0.8278359100619139,
"eval_loss": 0.44692954421043396,
"eval_macro avg": {
"f1-score": 0.7645361694310869,
"precision": 0.7833122850343401,
"recall": 0.7507151371753248,
"support": 27619.0
},
"eval_runtime": 4.7439,
"eval_samples_per_second": 16.864,
"eval_steps_per_second": 2.108,
"eval_weighted avg": {
"f1-score": 0.8236335905447046,
"precision": 0.8242076985653165,
"recall": 0.8278359100619139,
"support": 27619.0
},
"step": 162
},
{
"epoch": 3.0,
"eval_Claim": {
"f1-score": 0.6218357913273168,
"precision": 0.5744204961366409,
"recall": 0.6777831094049904,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.7799868909766223,
"precision": 0.7360824742268042,
"recall": 0.8294609665427509,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9093637121422617,
"precision": 0.9330596419204014,
"recall": 0.8868415347929763,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.876420695504665,
"precision": 0.89797514556357,
"recall": 0.855876749772219,
"support": 12073.0
},
"eval_accuracy": 0.8372859263550454,
"eval_loss": 0.40934404730796814,
"eval_macro avg": {
"f1-score": 0.7969017724877165,
"precision": 0.7853844394618541,
"recall": 0.8124905901282342,
"support": 27619.0
},
"eval_runtime": 4.7941,
"eval_samples_per_second": 16.687,
"eval_steps_per_second": 2.086,
"eval_weighted avg": {
"f1-score": 0.8414917278933443,
"precision": 0.8482528803063183,
"recall": 0.8372859263550454,
"support": 27619.0
},
"step": 243
},
{
"epoch": 4.0,
"eval_Claim": {
"f1-score": 0.608644284729743,
"precision": 0.516112873601603,
"recall": 0.7416026871401151,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.788337924701561,
"precision": 0.7790381125226861,
"recall": 0.7978624535315985,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9068679925629612,
"precision": 0.9442684500762643,
"recall": 0.8723173639713853,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8528029247910863,
"precision": 0.8985600293497202,
"recall": 0.8114801623457302,
"support": 12073.0
},
"eval_accuracy": 0.8201962417176581,
"eval_loss": 0.5160673260688782,
"eval_macro avg": {
"f1-score": 0.789163281696338,
"precision": 0.7844948663875684,
"recall": 0.8058156667472073,
"support": 27619.0
},
"eval_runtime": 4.7669,
"eval_samples_per_second": 16.782,
"eval_steps_per_second": 2.098,
"eval_weighted avg": {
"f1-score": 0.8289940404467937,
"precision": 0.8468005514342695,
"recall": 0.8201962417176581,
"support": 27619.0
},
"step": 324
},
{
"epoch": 5.0,
"eval_Claim": {
"f1-score": 0.5923049770561243,
"precision": 0.5811590856615101,
"recall": 0.6038867562380038,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.7829807463750891,
"precision": 0.8014598540145985,
"recall": 0.7653345724907064,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9075350077795066,
"precision": 0.9311288483466362,
"recall": 0.8851073054411446,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8770785784153896,
"precision": 0.8633555323758325,
"recall": 0.891244926695933,
"support": 12073.0
},
"eval_accuracy": 0.8360186827908324,
"eval_loss": 0.5153732299804688,
"eval_macro avg": {
"f1-score": 0.7899748274065275,
"precision": 0.7942758300996444,
"recall": 0.7863933902164468,
"support": 27619.0
},
"eval_runtime": 4.7506,
"eval_samples_per_second": 16.84,
"eval_steps_per_second": 2.105,
"eval_weighted avg": {
"f1-score": 0.8369451960444566,
"precision": 0.8385857117236715,
"recall": 0.8360186827908324,
"support": 27619.0
},
"step": 405
},
{
"epoch": 6.0,
"eval_Claim": {
"f1-score": 0.5533534584591154,
"precision": 0.6108374384236454,
"recall": 0.5057581573896354,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.7910518800571156,
"precision": 0.8107317073170732,
"recall": 0.7723048327137546,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.909817664129661,
"precision": 0.919331636605068,
"recall": 0.9004985909386516,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8747714081259442,
"precision": 0.8410671966975002,
"recall": 0.9112896546011762,
"support": 12073.0
},
"eval_accuracy": 0.8356566132010572,
"eval_loss": 0.5417770147323608,
"eval_macro avg": {
"f1-score": 0.782248602692959,
"precision": 0.7954919947608217,
"recall": 0.7724628089108043,
"support": 27619.0
},
"eval_runtime": 4.7432,
"eval_samples_per_second": 16.866,
"eval_steps_per_second": 2.108,
"eval_weighted avg": {
"f1-score": 0.8314498656832355,
"precision": 0.8301032992701898,
"recall": 0.8356566132010572,
"support": 27619.0
},
"step": 486
},
{
"epoch": 6.17,
"grad_norm": 6.106111526489258,
"learning_rate": 1.7530864197530865e-05,
"loss": 0.345,
"step": 500
},
{
"epoch": 7.0,
"eval_Claim": {
"f1-score": 0.5860923076923077,
"precision": 0.6017184735911044,
"recall": 0.571257197696737,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.8099526066350711,
"precision": 0.8264023210831721,
"recall": 0.7941449814126395,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9184632841627491,
"precision": 0.9157418381639909,
"recall": 0.9212009538261435,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8751742803247764,
"precision": 0.8666450093397222,
"recall": 0.8838731052762362,
"support": 12073.0
},
"eval_accuracy": 0.84217386581701,
"eval_loss": 0.5831862688064575,
"eval_macro avg": {
"f1-score": 0.7974206197037261,
"precision": 0.8026269105444974,
"recall": 0.792619059552939,
"support": 27619.0
},
"eval_runtime": 4.7225,
"eval_samples_per_second": 16.94,
"eval_steps_per_second": 2.118,
"eval_weighted avg": {
"f1-score": 0.8409273360363069,
"precision": 0.8399297508801243,
"recall": 0.84217386581701,
"support": 27619.0
},
"step": 567
},
{
"epoch": 8.0,
"eval_Claim": {
"f1-score": 0.6089751514402275,
"precision": 0.6281560826319816,
"recall": 0.5909309021113244,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.8056783802653014,
"precision": 0.806993006993007,
"recall": 0.804368029739777,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9201371876531113,
"precision": 0.9243136825987094,
"recall": 0.9159982657706481,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8842053669893396,
"precision": 0.8721998388396455,
"recall": 0.8965460117617825,
"support": 12073.0
},
"eval_accuracy": 0.8497411202433107,
"eval_loss": 0.5915889739990234,
"eval_macro avg": {
"f1-score": 0.8047490215869949,
"precision": 0.8079156527658359,
"recall": 0.801960802345883,
"support": 27619.0
},
"eval_runtime": 4.7226,
"eval_samples_per_second": 16.94,
"eval_steps_per_second": 2.117,
"eval_weighted avg": {
"f1-score": 0.8485544514458778,
"precision": 0.8476986926907486,
"recall": 0.8497411202433107,
"support": 27619.0
},
"step": 648
},
{
"epoch": 9.0,
"eval_Claim": {
"f1-score": 0.6137689614935823,
"precision": 0.5974557019536574,
"recall": 0.6309980806142035,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.8104347826086956,
"precision": 0.761437908496732,
"recall": 0.8661710037174721,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9178336980306346,
"precision": 0.9265518003092555,
"recall": 0.9092781270323,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8752312090129477,
"precision": 0.8886043533930857,
"recall": 0.8622546177420691,
"support": 12073.0
},
"eval_accuracy": 0.8433686954632681,
"eval_loss": 0.6680669784545898,
"eval_macro avg": {
"f1-score": 0.804317162786465,
"precision": 0.7935124410381827,
"recall": 0.8171754572765112,
"support": 27619.0
},
"eval_runtime": 4.7195,
"eval_samples_per_second": 16.951,
"eval_steps_per_second": 2.119,
"eval_weighted avg": {
"f1-score": 0.8449561811840803,
"precision": 0.8474346288061019,
"recall": 0.8433686954632681,
"support": 27619.0
},
"step": 729
},
{
"epoch": 10.0,
"eval_Claim": {
"f1-score": 0.5938368860055606,
"precision": 0.5741487455197133,
"recall": 0.6149232245681382,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.8202093764223942,
"precision": 0.8037466547725245,
"recall": 0.837360594795539,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9145026408450704,
"precision": 0.9286033519553073,
"recall": 0.9008237589421201,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8716092527874855,
"precision": 0.8756164841594918,
"recall": 0.8676385322620724,
"support": 12073.0
},
"eval_accuracy": 0.8382273072884608,
"eval_loss": 0.7243372201919556,
"eval_macro avg": {
"f1-score": 0.8000395390151277,
"precision": 0.7955288091017592,
"recall": 0.8051865276419675,
"support": 27619.0
},
"eval_runtime": 4.7291,
"eval_samples_per_second": 16.917,
"eval_steps_per_second": 2.115,
"eval_weighted avg": {
"f1-score": 0.840013852502701,
"precision": 0.8422219164630815,
"recall": 0.8382273072884608,
"support": 27619.0
},
"step": 810
},
{
"epoch": 11.0,
"eval_Claim": {
"f1-score": 0.6018801410105759,
"precision": 0.5898203592814372,
"recall": 0.614443378119002,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.7984663311766117,
"precision": 0.8243443839683325,
"recall": 0.7741635687732342,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9056203605514316,
"precision": 0.9334944195144402,
"recall": 0.8793626707132018,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.878886273236464,
"precision": 0.8616792678074016,
"recall": 0.8967945001242442,
"support": 12073.0
},
"eval_accuracy": 0.838806618632101,
"eval_loss": 0.7492838501930237,
"eval_macro avg": {
"f1-score": 0.7962132764937708,
"precision": 0.8023346076429029,
"recall": 0.7911910294324205,
"support": 27619.0
},
"eval_runtime": 4.7534,
"eval_samples_per_second": 16.83,
"eval_steps_per_second": 2.104,
"eval_weighted avg": {
"f1-score": 0.8397473983726959,
"precision": 0.8417333606018996,
"recall": 0.838806618632101,
"support": 27619.0
},
"step": 891
},
{
"epoch": 12.0,
"eval_Claim": {
"f1-score": 0.6351054481546573,
"precision": 0.5856969205834684,
"recall": 0.6936180422264875,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.8096330275229358,
"precision": 0.7993659420289855,
"recall": 0.8201672862453532,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9158836689038031,
"precision": 0.9461520684076727,
"recall": 0.8874918707999133,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8832342847576798,
"precision": 0.8926486760849336,
"recall": 0.8740164002319225,
"support": 12073.0
},
"eval_accuracy": 0.8470980122379521,
"eval_loss": 0.7472469806671143,
"eval_macro avg": {
"f1-score": 0.810964107334769,
"precision": 0.8059659017762651,
"recall": 0.8188233998759191,
"support": 27619.0
},
"eval_runtime": 4.7526,
"eval_samples_per_second": 16.833,
"eval_steps_per_second": 2.104,
"eval_weighted avg": {
"f1-score": 0.8509605717920246,
"precision": 0.8569306173916821,
"recall": 0.8470980122379521,
"support": 27619.0
},
"step": 972
},
{
"epoch": 12.35,
"grad_norm": 3.1761698722839355,
"learning_rate": 1.506172839506173e-05,
"loss": 0.0604,
"step": 1000
},
{
"epoch": 13.0,
"eval_Claim": {
"f1-score": 0.6302578018995931,
"precision": 0.5960222412318221,
"recall": 0.6686660268714012,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.8086405259450574,
"precision": 0.8172757475083057,
"recall": 0.800185873605948,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9224348987313599,
"precision": 0.9477475417333638,
"recall": 0.8984391935833514,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8844928196002152,
"precision": 0.8838709677419355,
"recall": 0.8851155470885447,
"support": 12073.0
},
"eval_accuracy": 0.8502842246279735,
"eval_loss": 0.812764048576355,
"eval_macro avg": {
"f1-score": 0.8114565115440564,
"precision": 0.8112291245538568,
"recall": 0.8131016602873112,
"support": 27619.0
},
"eval_runtime": 4.7412,
"eval_samples_per_second": 16.873,
"eval_steps_per_second": 2.109,
"eval_weighted avg": {
"f1-score": 0.8528902247322565,
"precision": 0.8565802934093382,
"recall": 0.8502842246279735,
"support": 27619.0
},
"step": 1053
},
{
"epoch": 14.0,
"eval_Claim": {
"f1-score": 0.6177896267571498,
"precision": 0.6241429970617042,
"recall": 0.6115642994241842,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.7891914191419142,
"precision": 0.7095697329376854,
"recall": 0.8889405204460966,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.915085817524842,
"precision": 0.9549846806504831,
"recall": 0.8783871667027965,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8833210513387375,
"precision": 0.8733101270946328,
"recall": 0.8935641514122422,
"support": 12073.0
},
"eval_accuracy": 0.8455773199608965,
"eval_loss": 0.8994765281677246,
"eval_macro avg": {
"f1-score": 0.8013469786906608,
"precision": 0.7905018844361265,
"recall": 0.8181140344963299,
"support": 27619.0
},
"eval_runtime": 4.7538,
"eval_samples_per_second": 16.829,
"eval_steps_per_second": 2.104,
"eval_weighted avg": {
"f1-score": 0.8465260836240982,
"precision": 0.850232952139105,
"recall": 0.8455773199608965,
"support": 27619.0
},
"step": 1134
},
{
"epoch": 15.0,
"eval_Claim": {
"f1-score": 0.6071797224452345,
"precision": 0.5816304108987036,
"recall": 0.6350767754318618,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.8077178975382568,
"precision": 0.7725922783198982,
"recall": 0.8461895910780669,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9198298062926884,
"precision": 0.9512505789717461,
"recall": 0.8904183828311294,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8720390922643698,
"precision": 0.8719668737060041,
"recall": 0.8721113227863828,
"support": 12073.0
},
"eval_accuracy": 0.8404359317860893,
"eval_loss": 0.9128327369689941,
"eval_macro avg": {
"f1-score": 0.8016916296351374,
"precision": 0.794360035474088,
"recall": 0.8109490180318603,
"support": 27619.0
},
"eval_runtime": 4.7243,
"eval_samples_per_second": 16.934,
"eval_steps_per_second": 2.117,
"eval_weighted avg": {
"f1-score": 0.8430215341764057,
"precision": 0.8468933720777774,
"recall": 0.8404359317860893,
"support": 27619.0
},
"step": 1215
},
{
"epoch": 16.0,
"eval_Claim": {
"f1-score": 0.6309481002037637,
"precision": 0.6304191616766467,
"recall": 0.6314779270633397,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.8244803695150116,
"precision": 0.8195592286501377,
"recall": 0.8294609665427509,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9188405797101449,
"precision": 0.9101446193109315,
"recall": 0.9277043138955127,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8854815124294965,
"precision": 0.8933569381217332,
"recall": 0.8777437256688478,
"support": 12073.0
},
"eval_accuracy": 0.8535066439769724,
"eval_loss": 0.8802785873413086,
"eval_macro avg": {
"f1-score": 0.8149376404646043,
"precision": 0.8133699869398623,
"recall": 0.8165967332926127,
"support": 27619.0
},
"eval_runtime": 4.7225,
"eval_samples_per_second": 16.94,
"eval_steps_per_second": 2.118,
"eval_weighted avg": {
"f1-score": 0.8534601153123827,
"precision": 0.8535345630772185,
"recall": 0.8535066439769724,
"support": 27619.0
},
"step": 1296
},
{
"epoch": 17.0,
"eval_Claim": {
"f1-score": 0.6351891003053795,
"precision": 0.6221813161527842,
"recall": 0.6487523992322457,
"support": 4168.0
},
"eval_MajorClaim": {
"f1-score": 0.8157058556513844,
"precision": 0.7972493345164152,
"recall": 0.8350371747211895,
"support": 2152.0
},
"eval_O": {
"f1-score": 0.9181362943609842,
"precision": 0.9221517603323858,
"recall": 0.914155647084327,
"support": 9226.0
},
"eval_Premise": {
"f1-score": 0.8847406664996242,
"precision": 0.8921923692411353,
"recall": 0.8774124078522323,
"support": 12073.0
},
"eval_accuracy": 0.8518773308229842,
"eval_loss": 0.8617361783981323,
"eval_macro avg": {
"f1-score": 0.813442979204343,
"precision": 0.8084436950606801,
"recall": 0.8188394072224987,
"support": 27619.0
},
"eval_runtime": 4.7584,
"eval_samples_per_second": 16.812,
"eval_steps_per_second": 2.102,
"eval_weighted avg": {
"f1-score": 0.8528573333523664,
"precision": 0.8540549226358285,
"recall": 0.8518773308229842,
"support": 27619.0
},
"step": 1377
}
],
"logging_steps": 500,
"max_steps": 4050,
"num_input_tokens_seen": 0,
"num_train_epochs": 50,
"save_steps": 500,
"total_flos": 2444408456355600.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}