| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 729, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.0, | |
| "precision": 0.0, | |
| "recall": 0.0, | |
| "support": 271.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.0, | |
| "precision": 0.0, | |
| "recall": 0.0, | |
| "support": 139.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.6746310611384398, | |
| "precision": 0.6075949367088608, | |
| "recall": 0.7582938388625592, | |
| "support": 633.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.4376506024096386, | |
| "precision": 0.550587343690792, | |
| "recall": 0.3631592101974506, | |
| "support": 4001.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.620034542314335, | |
| "precision": 0.5483008781977854, | |
| "recall": 0.7133631395926477, | |
| "support": 2013.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8620734063103671, | |
| "precision": 0.8396186972154862, | |
| "recall": 0.8857621736062103, | |
| "support": 11336.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.8807729058286443, | |
| "precision": 0.8630878069080317, | |
| "recall": 0.8991979189247779, | |
| "support": 9226.0 | |
| }, | |
| "eval_accuracy": 0.785908251565951, | |
| "eval_loss": 0.5655931830406189, | |
| "eval_macro avg": { | |
| "f1-score": 0.49645178828591785, | |
| "precision": 0.48702709467442235, | |
| "recall": 0.5171108973119495, | |
| "support": 27619.0 | |
| }, | |
| "eval_runtime": 4.8252, | |
| "eval_samples_per_second": 16.58, | |
| "eval_steps_per_second": 2.072, | |
| "eval_weighted avg": { | |
| "f1-score": 0.7721020318885458, | |
| "precision": 0.766573839857488, | |
| "recall": 0.785908251565951, | |
| "support": 27619.0 | |
| }, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.36036036036036034, | |
| "precision": 0.4624277456647399, | |
| "recall": 0.2952029520295203, | |
| "support": 271.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.5836909871244635, | |
| "precision": 0.723404255319149, | |
| "recall": 0.4892086330935252, | |
| "support": 139.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7396491228070177, | |
| "precision": 0.6654040404040404, | |
| "recall": 0.8325434439178515, | |
| "support": 633.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.46704216775764956, | |
| "precision": 0.5973520249221184, | |
| "recall": 0.3834041489627593, | |
| "support": 4001.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7326456649010316, | |
| "precision": 0.8348157560355781, | |
| "recall": 0.6527570789865872, | |
| "support": 2013.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8679505932065659, | |
| "precision": 0.8045344983428744, | |
| "recall": 0.9422194777699365, | |
| "support": 11336.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9101698606271778, | |
| "precision": 0.9143513454386348, | |
| "recall": 0.9060264469976155, | |
| "support": 9226.0 | |
| }, | |
| "eval_accuracy": 0.8169376154096818, | |
| "eval_loss": 0.4831336438655853, | |
| "eval_macro avg": { | |
| "f1-score": 0.6659298223977524, | |
| "precision": 0.7146128094467336, | |
| "recall": 0.6430517402511137, | |
| "support": 27619.0 | |
| }, | |
| "eval_runtime": 4.8125, | |
| "eval_samples_per_second": 16.623, | |
| "eval_steps_per_second": 2.078, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8047632099274016, | |
| "precision": 0.8064582361050344, | |
| "recall": 0.8169376154096818, | |
| "support": 27619.0 | |
| }, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5806451612903225, | |
| "precision": 0.59765625, | |
| "recall": 0.5645756457564576, | |
| "support": 271.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7138047138047138, | |
| "precision": 0.6708860759493671, | |
| "recall": 0.762589928057554, | |
| "support": 139.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7739984882842026, | |
| "precision": 0.7420289855072464, | |
| "recall": 0.8088467614533965, | |
| "support": 633.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.6128912532475566, | |
| "precision": 0.6068103870651641, | |
| "recall": 0.6190952261934516, | |
| "support": 4001.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.8072942336126171, | |
| "precision": 0.8009779951100244, | |
| "recall": 0.8137108792846498, | |
| "support": 2013.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8797559224694903, | |
| "precision": 0.8951789627465303, | |
| "recall": 0.8648553281580804, | |
| "support": 11336.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9226235130211126, | |
| "precision": 0.912356930902925, | |
| "recall": 0.933123780619987, | |
| "support": 9226.0 | |
| }, | |
| "eval_accuracy": 0.8435859372171332, | |
| "eval_loss": 0.4283505082130432, | |
| "eval_macro avg": { | |
| "f1-score": 0.7558590408185736, | |
| "precision": 0.7465565124687511, | |
| "recall": 0.7666853642176539, | |
| "support": 27619.0 | |
| }, | |
| "eval_runtime": 4.8554, | |
| "eval_samples_per_second": 16.476, | |
| "eval_steps_per_second": 2.06, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8439412578936614, | |
| "precision": 0.8447189682878564, | |
| "recall": 0.8435859372171332, | |
| "support": 27619.0 | |
| }, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.6033333333333333, | |
| "precision": 0.5501519756838906, | |
| "recall": 0.6678966789667896, | |
| "support": 271.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.6847457627118644, | |
| "precision": 0.6474358974358975, | |
| "recall": 0.7266187050359713, | |
| "support": 139.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7818471337579618, | |
| "precision": 0.7881219903691814, | |
| "recall": 0.7756714060031595, | |
| "support": 633.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.6216657593903103, | |
| "precision": 0.550733024691358, | |
| "recall": 0.7135716070982254, | |
| "support": 4001.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7652218782249742, | |
| "precision": 0.7960279119699409, | |
| "recall": 0.7367113760556383, | |
| "support": 2013.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8739165400768475, | |
| "precision": 0.8853883758826725, | |
| "recall": 0.8627381792519407, | |
| "support": 11336.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9062570845613239, | |
| "precision": 0.9497505345687812, | |
| "recall": 0.8665727292434424, | |
| "support": 9226.0 | |
| }, | |
| "eval_accuracy": 0.8286324631594192, | |
| "eval_loss": 0.4826153814792633, | |
| "eval_macro avg": { | |
| "f1-score": 0.7481410702938023, | |
| "precision": 0.7382299586573888, | |
| "recall": 0.7642543830935953, | |
| "support": 27619.0 | |
| }, | |
| "eval_runtime": 4.8516, | |
| "eval_samples_per_second": 16.49, | |
| "eval_steps_per_second": 2.061, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8345383371838407, | |
| "precision": 0.8451795530099128, | |
| "recall": 0.8286324631594192, | |
| "support": 27619.0 | |
| }, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5825242718446603, | |
| "precision": 0.6147540983606558, | |
| "recall": 0.5535055350553506, | |
| "support": 271.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7317073170731707, | |
| "precision": 0.7094594594594594, | |
| "recall": 0.7553956834532374, | |
| "support": 139.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7840735068912711, | |
| "precision": 0.7607726597325408, | |
| "recall": 0.8088467614533965, | |
| "support": 633.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5830055074744296, | |
| "precision": 0.6132413793103448, | |
| "recall": 0.5556110972256936, | |
| "support": 4001.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.803728638011393, | |
| "precision": 0.8393726338561385, | |
| "recall": 0.7709885742672627, | |
| "support": 2013.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8823829787234042, | |
| "precision": 0.8523512002630713, | |
| "recall": 0.9146083274523642, | |
| "support": 11336.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9254767941792524, | |
| "precision": 0.9415657245401525, | |
| "recall": 0.9099284630392369, | |
| "support": 9226.0 | |
| }, | |
| "eval_accuracy": 0.8438031789709982, | |
| "eval_loss": 0.513404905796051, | |
| "eval_macro avg": { | |
| "f1-score": 0.7561284305996544, | |
| "precision": 0.7616453079317661, | |
| "recall": 0.7526977774209346, | |
| "support": 27619.0 | |
| }, | |
| "eval_runtime": 4.8184, | |
| "eval_samples_per_second": 16.603, | |
| "eval_steps_per_second": 2.075, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8417228378374912, | |
| "precision": 0.8414191958613282, | |
| "recall": 0.8438031789709982, | |
| "support": 27619.0 | |
| }, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5851851851851851, | |
| "precision": 0.587360594795539, | |
| "recall": 0.5830258302583026, | |
| "support": 271.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7460815047021944, | |
| "precision": 0.6611111111111111, | |
| "recall": 0.8561151079136691, | |
| "support": 139.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7778631821675633, | |
| "precision": 0.7574850299401198, | |
| "recall": 0.7993680884676145, | |
| "support": 633.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5999495204442201, | |
| "precision": 0.6059138414478715, | |
| "recall": 0.5941014746313422, | |
| "support": 4001.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.800587227795449, | |
| "precision": 0.7888138862102217, | |
| "recall": 0.8127173373075013, | |
| "support": 2013.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8798354629791703, | |
| "precision": 0.8729593608891977, | |
| "recall": 0.8868207480592801, | |
| "support": 11336.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9269283557507548, | |
| "precision": 0.9391478473690066, | |
| "recall": 0.9150227617602428, | |
| "support": 9226.0 | |
| }, | |
| "eval_accuracy": 0.843296281545313, | |
| "eval_loss": 0.5017187595367432, | |
| "eval_macro avg": { | |
| "f1-score": 0.7594900627177908, | |
| "precision": 0.744684524537581, | |
| "recall": 0.7781673354854218, | |
| "support": 27619.0 | |
| }, | |
| "eval_runtime": 4.8298, | |
| "eval_samples_per_second": 16.564, | |
| "eval_steps_per_second": 2.07, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8433438519855446, | |
| "precision": 0.8437360576786502, | |
| "recall": 0.843296281545313, | |
| "support": 27619.0 | |
| }, | |
| "step": 486 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "grad_norm": 7.375185489654541, | |
| "learning_rate": 1.7530864197530865e-05, | |
| "loss": 0.4085, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5725338491295938, | |
| "precision": 0.6016260162601627, | |
| "recall": 0.5461254612546126, | |
| "support": 271.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.745644599303136, | |
| "precision": 0.722972972972973, | |
| "recall": 0.7697841726618705, | |
| "support": 139.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7915087187263078, | |
| "precision": 0.760932944606414, | |
| "recall": 0.8246445497630331, | |
| "support": 633.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5842240562846706, | |
| "precision": 0.6368731563421829, | |
| "recall": 0.5396150962259435, | |
| "support": 4001.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7921810699588476, | |
| "precision": 0.8213333333333334, | |
| "recall": 0.7650273224043715, | |
| "support": 2013.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8817056396148556, | |
| "precision": 0.8598256203890007, | |
| "recall": 0.9047282992237121, | |
| "support": 11336.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9274176179194485, | |
| "precision": 0.9214637277979885, | |
| "recall": 0.9334489486234554, | |
| "support": 9226.0 | |
| }, | |
| "eval_accuracy": 0.8452152503711213, | |
| "eval_loss": 0.6029371023178101, | |
| "eval_macro avg": { | |
| "f1-score": 0.7564593644195513, | |
| "precision": 0.7607182531002935, | |
| "recall": 0.7547676928795714, | |
| "support": 27619.0 | |
| }, | |
| "eval_runtime": 4.8043, | |
| "eval_samples_per_second": 16.652, | |
| "eval_steps_per_second": 2.081, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8415705604110203, | |
| "precision": 0.8398235103191511, | |
| "recall": 0.8452152503711213, | |
| "support": 27619.0 | |
| }, | |
| "step": 567 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5641891891891891, | |
| "precision": 0.5202492211838006, | |
| "recall": 0.6162361623616236, | |
| "support": 271.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7318840579710144, | |
| "precision": 0.7372262773722628, | |
| "recall": 0.7266187050359713, | |
| "support": 139.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7673716012084594, | |
| "precision": 0.7351664254703328, | |
| "recall": 0.8025276461295419, | |
| "support": 633.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5930426495115559, | |
| "precision": 0.5665832005463237, | |
| "recall": 0.6220944763809048, | |
| "support": 4001.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7816030123722432, | |
| "precision": 0.852199413489736, | |
| "recall": 0.7218082463984103, | |
| "support": 2013.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8746564286025915, | |
| "precision": 0.8652567975830816, | |
| "recall": 0.8842625264643613, | |
| "support": 11336.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.924221395658691, | |
| "precision": 0.9473085239558439, | |
| "recall": 0.9022328202904835, | |
| "support": 9226.0 | |
| }, | |
| "eval_accuracy": 0.835149715775372, | |
| "eval_loss": 0.6411188840866089, | |
| "eval_macro avg": { | |
| "f1-score": 0.7481383335019636, | |
| "precision": 0.7462842656573401, | |
| "recall": 0.7536829404373281, | |
| "support": 27619.0 | |
| }, | |
| "eval_runtime": 4.8296, | |
| "eval_samples_per_second": 16.564, | |
| "eval_steps_per_second": 2.071, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8374117728187233, | |
| "precision": 0.8414359188593861, | |
| "recall": 0.835149715775372, | |
| "support": 27619.0 | |
| }, | |
| "step": 648 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.6241610738255032, | |
| "precision": 0.5723076923076923, | |
| "recall": 0.6863468634686347, | |
| "support": 271.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.756578947368421, | |
| "precision": 0.696969696969697, | |
| "recall": 0.8273381294964028, | |
| "support": 139.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.780448717948718, | |
| "precision": 0.791869918699187, | |
| "recall": 0.7693522906793049, | |
| "support": 633.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.640754369825207, | |
| "precision": 0.5933972310969116, | |
| "recall": 0.69632591852037, | |
| "support": 4001.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.797244094488189, | |
| "precision": 0.7898586055582643, | |
| "recall": 0.8047690014903129, | |
| "support": 2013.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8789423814465549, | |
| "precision": 0.8996028447400019, | |
| "recall": 0.8592095977417078, | |
| "support": 11336.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9287169042769858, | |
| "precision": 0.9435186220780674, | |
| "recall": 0.9143724257533059, | |
| "support": 9226.0 | |
| }, | |
| "eval_accuracy": 0.8461566313045368, | |
| "eval_loss": 0.6282346844673157, | |
| "eval_macro avg": { | |
| "f1-score": 0.7724066413113684, | |
| "precision": 0.7553606587785459, | |
| "recall": 0.7939591753071484, | |
| "support": 27619.0 | |
| }, | |
| "eval_runtime": 4.8069, | |
| "eval_samples_per_second": 16.643, | |
| "eval_steps_per_second": 2.08, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8497367665000708, | |
| "precision": 0.85521547416283, | |
| "recall": 0.8461566313045368, | |
| "support": 27619.0 | |
| }, | |
| "step": 729 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4050, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1294117314462000.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |