| { | |
| "best_metric": 0.508861243724823, | |
| "best_model_checkpoint": "longformer-full_labels/checkpoint-324", | |
| "epoch": 19.0, | |
| "eval_steps": 500, | |
| "global_step": 1539, | |
| "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": 0.006993006993006993, | |
| "precision": 0.5, | |
| "recall": 0.0035211267605633804, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.0, | |
| "precision": 0.0, | |
| "recall": 0.0, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.6336375488917861, | |
| "precision": 0.5883777239709443, | |
| "recall": 0.6864406779661016, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.37228850325379614, | |
| "precision": 0.41618672324946954, | |
| "recall": 0.3367672308069659, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.49153013051930017, | |
| "precision": 0.5611921369689283, | |
| "recall": 0.43725296442687744, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.846019669697532, | |
| "precision": 0.7746890504995582, | |
| "recall": 0.9318181818181818, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.8650519031141869, | |
| "precision": 0.9111808904340025, | |
| "recall": 0.8233684637211187, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.7591191109292971, | |
| "eval_loss": 0.6442636251449585, | |
| "eval_macro avg": { | |
| "f1-score": 0.4593601089242298, | |
| "precision": 0.535946646446129, | |
| "recall": 0.4598812350714013, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.1153, | |
| "eval_samples_per_second": 13.082, | |
| "eval_steps_per_second": 1.635, | |
| "eval_weighted avg": { | |
| "f1-score": 0.7448054609057474, | |
| "precision": 0.7451676238152983, | |
| "recall": 0.7591191109292971, | |
| "support": 29334.0 | |
| }, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.47385620915032683, | |
| "precision": 0.4420731707317073, | |
| "recall": 0.5105633802816901, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.5267489711934156, | |
| "precision": 0.6274509803921569, | |
| "recall": 0.45390070921985815, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7686170212765958, | |
| "precision": 0.7261306532663316, | |
| "recall": 0.8163841807909604, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5068787075602675, | |
| "precision": 0.5221008840353614, | |
| "recall": 0.49251900907530044, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.6644462947543713, | |
| "precision": 0.5741007194244604, | |
| "recall": 0.7885375494071146, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8735948241002829, | |
| "precision": 0.8642982877260361, | |
| "recall": 0.8830935251798561, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.8825588796944621, | |
| "precision": 0.925979519145147, | |
| "recall": 0.8430279691933522, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8015272380173177, | |
| "eval_loss": 0.5138605833053589, | |
| "eval_macro avg": { | |
| "f1-score": 0.6709572725328175, | |
| "precision": 0.668876316388743, | |
| "recall": 0.6840037604497332, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.1482, | |
| "eval_samples_per_second": 13.012, | |
| "eval_steps_per_second": 1.627, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8031401896750007, | |
| "precision": 0.8089032379475054, | |
| "recall": 0.8015272380173177, | |
| "support": 29334.0 | |
| }, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.4233576642335767, | |
| "precision": 0.4393939393939394, | |
| "recall": 0.4084507042253521, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.6294820717131473, | |
| "precision": 0.7181818181818181, | |
| "recall": 0.5602836879432624, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7608562691131499, | |
| "precision": 0.6709816612729234, | |
| "recall": 0.8785310734463276, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.46000546000546, | |
| "precision": 0.5186211141889813, | |
| "recall": 0.41329408879077756, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.6850351161534306, | |
| "precision": 0.7556615017878426, | |
| "recall": 0.6264822134387352, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8689253296477533, | |
| "precision": 0.7999862438957287, | |
| "recall": 0.9508665794637018, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.8740981828044481, | |
| "precision": 0.9404692424419283, | |
| "recall": 0.8164775030401297, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.7996863707643008, | |
| "eval_loss": 0.5769864320755005, | |
| "eval_macro avg": { | |
| "f1-score": 0.6716800133815666, | |
| "precision": 0.6918993601661659, | |
| "recall": 0.6649122643354695, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0693, | |
| "eval_samples_per_second": 13.181, | |
| "eval_steps_per_second": 1.648, | |
| "eval_weighted avg": { | |
| "f1-score": 0.7930703491848509, | |
| "precision": 0.7980829724295806, | |
| "recall": 0.7996863707643008, | |
| "support": 29334.0 | |
| }, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5486211901306242, | |
| "precision": 0.4666666666666667, | |
| "recall": 0.6654929577464789, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.6866666666666668, | |
| "precision": 0.6477987421383647, | |
| "recall": 0.7304964539007093, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7603550295857988, | |
| "precision": 0.7981366459627329, | |
| "recall": 0.7259887005649718, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5798045602605862, | |
| "precision": 0.5201636469900643, | |
| "recall": 0.6548933038999264, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7716646989374262, | |
| "precision": 0.7390321121664405, | |
| "recall": 0.8073122529644269, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8704094554664417, | |
| "precision": 0.899807994414383, | |
| "recall": 0.842871157619359, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.8969362234264276, | |
| "precision": 0.9231016731016731, | |
| "recall": 0.8722132144304824, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8190836571896093, | |
| "eval_loss": 0.508861243724823, | |
| "eval_macro avg": { | |
| "f1-score": 0.7306368320677102, | |
| "precision": 0.7135296402057607, | |
| "recall": 0.7570382915894793, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0931, | |
| "eval_samples_per_second": 13.13, | |
| "eval_steps_per_second": 1.641, | |
| "eval_weighted avg": { | |
| "f1-score": 0.825475128990697, | |
| "precision": 0.835926930625345, | |
| "recall": 0.8190836571896093, | |
| "support": 29334.0 | |
| }, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5466237942122186, | |
| "precision": 0.5029585798816568, | |
| "recall": 0.5985915492957746, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.6842105263157895, | |
| "precision": 0.728, | |
| "recall": 0.6453900709219859, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.775623268698061, | |
| "precision": 0.7608695652173914, | |
| "recall": 0.7909604519774012, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5628513335725392, | |
| "precision": 0.5492530345471522, | |
| "recall": 0.577140053961246, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7028394714647174, | |
| "precision": 0.8153946510110893, | |
| "recall": 0.6175889328063241, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8796072750684049, | |
| "precision": 0.8660855784469097, | |
| "recall": 0.8935578809679529, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.8964530307676581, | |
| "precision": 0.9043101670447515, | |
| "recall": 0.8887312525334414, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8223904002181769, | |
| "eval_loss": 0.5749920606613159, | |
| "eval_macro avg": { | |
| "f1-score": 0.7211726714427698, | |
| "precision": 0.7324102251641358, | |
| "recall": 0.715994313209161, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0798, | |
| "eval_samples_per_second": 13.158, | |
| "eval_steps_per_second": 1.645, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8223802682715222, | |
| "precision": 0.8246928072651426, | |
| "recall": 0.8223904002181769, | |
| "support": 29334.0 | |
| }, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5645933014354066, | |
| "precision": 0.5160349854227405, | |
| "recall": 0.6232394366197183, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7272727272727273, | |
| "precision": 0.6923076923076923, | |
| "recall": 0.7659574468085106, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7851644506648006, | |
| "precision": 0.7780859916782247, | |
| "recall": 0.7923728813559322, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5951509606587375, | |
| "precision": 0.5575316048853654, | |
| "recall": 0.6382143733137111, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7690405539070228, | |
| "precision": 0.7698019801980198, | |
| "recall": 0.7682806324110671, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8794871794871795, | |
| "precision": 0.8899397388684298, | |
| "recall": 0.8692773054283846, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.907427508140797, | |
| "precision": 0.9260470513767275, | |
| "recall": 0.8895419537900284, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8322765391695643, | |
| "eval_loss": 0.5503024458885193, | |
| "eval_macro avg": { | |
| "f1-score": 0.7468766687952387, | |
| "precision": 0.7328212921053143, | |
| "recall": 0.763840575675336, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0682, | |
| "eval_samples_per_second": 13.184, | |
| "eval_steps_per_second": 1.648, | |
| "eval_weighted avg": { | |
| "f1-score": 0.835690214793681, | |
| "precision": 0.8403274341189826, | |
| "recall": 0.8322765391695643, | |
| "support": 29334.0 | |
| }, | |
| "step": 486 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "grad_norm": 7.526494026184082, | |
| "learning_rate": 1.3827160493827162e-05, | |
| "loss": 0.4181, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5531914893617021, | |
| "precision": 0.5571428571428572, | |
| "recall": 0.5492957746478874, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7228070175438596, | |
| "precision": 0.7152777777777778, | |
| "recall": 0.7304964539007093, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7938420348058902, | |
| "precision": 0.7544529262086515, | |
| "recall": 0.8375706214689266, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5534057778955738, | |
| "precision": 0.6019025655808591, | |
| "recall": 0.5121412803532008, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7677873338545738, | |
| "precision": 0.8124655267512411, | |
| "recall": 0.7277667984189723, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8876130554463233, | |
| "precision": 0.855129565085619, | |
| "recall": 0.9226618705035972, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9097990979909799, | |
| "precision": 0.9203649937785151, | |
| "recall": 0.8994730441832185, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8377650507943001, | |
| "eval_loss": 0.6418657302856445, | |
| "eval_macro avg": { | |
| "f1-score": 0.7412065438427005, | |
| "precision": 0.7452480303322171, | |
| "recall": 0.7399151204966445, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.075, | |
| "eval_samples_per_second": 13.169, | |
| "eval_steps_per_second": 1.646, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8340655773672634, | |
| "precision": 0.8329491032454597, | |
| "recall": 0.8377650507943001, | |
| "support": 29334.0 | |
| }, | |
| "step": 567 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5652951699463328, | |
| "precision": 0.5745454545454546, | |
| "recall": 0.5563380281690141, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7197231833910034, | |
| "precision": 0.7027027027027027, | |
| "recall": 0.7375886524822695, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7941176470588234, | |
| "precision": 0.7538071065989848, | |
| "recall": 0.8389830508474576, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5740005295207837, | |
| "precision": 0.6235260281852172, | |
| "recall": 0.5317635516311013, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.773091849935317, | |
| "precision": 0.8115154807170016, | |
| "recall": 0.7381422924901185, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8920165582495565, | |
| "precision": 0.8614178024822965, | |
| "recall": 0.9248691955526488, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9097236438075741, | |
| "precision": 0.9189412737799835, | |
| "recall": 0.9006890960680989, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8426740301356788, | |
| "eval_loss": 0.6667934656143188, | |
| "eval_macro avg": { | |
| "f1-score": 0.7468526545584844, | |
| "precision": 0.7494936927159488, | |
| "recall": 0.7469105524629585, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0698, | |
| "eval_samples_per_second": 13.18, | |
| "eval_steps_per_second": 1.648, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8392138000943469, | |
| "precision": 0.8381245456177384, | |
| "recall": 0.8426740301356788, | |
| "support": 29334.0 | |
| }, | |
| "step": 648 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5863192182410424, | |
| "precision": 0.5454545454545454, | |
| "recall": 0.6338028169014085, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7210884353741497, | |
| "precision": 0.6928104575163399, | |
| "recall": 0.75177304964539, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7762879322512349, | |
| "precision": 0.7757404795486601, | |
| "recall": 0.7768361581920904, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.6114039290848108, | |
| "precision": 0.5975181456333412, | |
| "recall": 0.6259504537650233, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7696835908756438, | |
| "precision": 0.7642474427666829, | |
| "recall": 0.775197628458498, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8845693533077462, | |
| "precision": 0.893157763146929, | |
| "recall": 0.8761445389143231, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9076040026413369, | |
| "precision": 0.9098686220592729, | |
| "recall": 0.9053506282934739, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8388900252266994, | |
| "eval_loss": 0.7191675305366516, | |
| "eval_macro avg": { | |
| "f1-score": 0.7509937802537092, | |
| "precision": 0.7398282080179673, | |
| "recall": 0.7635793248814581, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0723, | |
| "eval_samples_per_second": 13.175, | |
| "eval_steps_per_second": 1.647, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8401384747371046, | |
| "precision": 0.8416318009865815, | |
| "recall": 0.8388900252266994, | |
| "support": 29334.0 | |
| }, | |
| "step": 729 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5391621129326047, | |
| "precision": 0.5584905660377358, | |
| "recall": 0.5211267605633803, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7254237288135594, | |
| "precision": 0.6948051948051948, | |
| "recall": 0.7588652482269503, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7871054398925452, | |
| "precision": 0.7503201024327785, | |
| "recall": 0.827683615819209, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5272531030760929, | |
| "precision": 0.5859070464767616, | |
| "recall": 0.4792739759627177, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7522123893805309, | |
| "precision": 0.7485322896281801, | |
| "recall": 0.7559288537549407, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8810152126668737, | |
| "precision": 0.8385786052009456, | |
| "recall": 0.92797580117724, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9004034159375491, | |
| "precision": 0.9320967566981234, | |
| "recall": 0.8707944872314552, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8273334696938706, | |
| "eval_loss": 0.8727766871452332, | |
| "eval_macro avg": { | |
| "f1-score": 0.7303679146713938, | |
| "precision": 0.7298186516113886, | |
| "recall": 0.7345212489622704, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.1014, | |
| "eval_samples_per_second": 13.112, | |
| "eval_steps_per_second": 1.639, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8231582874630071, | |
| "precision": 0.8231745470223255, | |
| "recall": 0.8273334696938706, | |
| "support": 29334.0 | |
| }, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5882352941176471, | |
| "precision": 0.5487804878048781, | |
| "recall": 0.6338028169014085, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7417218543046358, | |
| "precision": 0.6956521739130435, | |
| "recall": 0.7943262411347518, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7794221282593374, | |
| "precision": 0.7777777777777778, | |
| "recall": 0.7810734463276836, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.6072943172179812, | |
| "precision": 0.600095785440613, | |
| "recall": 0.6146676477802305, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7846607669616519, | |
| "precision": 0.7808219178082192, | |
| "recall": 0.7885375494071146, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8809899510007474, | |
| "precision": 0.8951898734177215, | |
| "recall": 0.8672334859385219, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9068834035721453, | |
| "precision": 0.8980524642289348, | |
| "recall": 0.9158897446291042, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8383786732119725, | |
| "eval_loss": 0.7904127836227417, | |
| "eval_macro avg": { | |
| "f1-score": 0.755601102204878, | |
| "precision": 0.742338640055884, | |
| "recall": 0.7707901331598307, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0781, | |
| "eval_samples_per_second": 13.162, | |
| "eval_steps_per_second": 1.645, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8390590885154817, | |
| "precision": 0.8401010980182351, | |
| "recall": 0.8383786732119725, | |
| "support": 29334.0 | |
| }, | |
| "step": 891 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5663082437275986, | |
| "precision": 0.5766423357664233, | |
| "recall": 0.5563380281690141, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7323943661971831, | |
| "precision": 0.7272727272727273, | |
| "recall": 0.7375886524822695, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.793821356615178, | |
| "precision": 0.7567221510883483, | |
| "recall": 0.8347457627118644, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5636018443178736, | |
| "precision": 0.6302699423718532, | |
| "recall": 0.5096884964434634, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7598984771573604, | |
| "precision": 0.7813152400835073, | |
| "recall": 0.7396245059288538, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.889289352033221, | |
| "precision": 0.8537686174213931, | |
| "recall": 0.9278940483976456, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9031132994819715, | |
| "precision": 0.9143213210094506, | |
| "recall": 0.892176732873936, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.838003681734506, | |
| "eval_loss": 0.9021136164665222, | |
| "eval_macro avg": { | |
| "f1-score": 0.7440609913614838, | |
| "precision": 0.7486160478591003, | |
| "recall": 0.7425794610010066, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0777, | |
| "eval_samples_per_second": 13.163, | |
| "eval_steps_per_second": 1.645, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8335608269497821, | |
| "precision": 0.8324430451309904, | |
| "recall": 0.838003681734506, | |
| "support": 29334.0 | |
| }, | |
| "step": 972 | |
| }, | |
| { | |
| "epoch": 12.35, | |
| "grad_norm": 1.9324438571929932, | |
| "learning_rate": 7.654320987654322e-06, | |
| "loss": 0.0774, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5774877650897227, | |
| "precision": 0.5379939209726444, | |
| "recall": 0.6232394366197183, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7087719298245613, | |
| "precision": 0.7013888888888888, | |
| "recall": 0.7163120567375887, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.784636488340192, | |
| "precision": 0.7626666666666667, | |
| "recall": 0.807909604519774, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5893326955273857, | |
| "precision": 0.5750291715285881, | |
| "recall": 0.6043659553593328, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7561601642710472, | |
| "precision": 0.7868589743589743, | |
| "recall": 0.7277667984189723, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8846897663174859, | |
| "precision": 0.8721798538290435, | |
| "recall": 0.8975637671680837, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.8962254341270668, | |
| "precision": 0.9202434336963485, | |
| "recall": 0.8734292663153628, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8312879252744256, | |
| "eval_loss": 0.9174407720565796, | |
| "eval_macro avg": { | |
| "f1-score": 0.7424720347853516, | |
| "precision": 0.7366229871344505, | |
| "recall": 0.7500838407341189, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0935, | |
| "eval_samples_per_second": 13.129, | |
| "eval_steps_per_second": 1.641, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8324170375280131, | |
| "precision": 0.8344622887797986, | |
| "recall": 0.8312879252744256, | |
| "support": 29334.0 | |
| }, | |
| "step": 1053 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5770491803278688, | |
| "precision": 0.5398773006134969, | |
| "recall": 0.6197183098591549, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.736842105263158, | |
| "precision": 0.6871165644171779, | |
| "recall": 0.7943262411347518, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7869535045107564, | |
| "precision": 0.7735334242837654, | |
| "recall": 0.8008474576271186, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5928150907124834, | |
| "precision": 0.5810174281676872, | |
| "recall": 0.6051017905322541, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7852270054104916, | |
| "precision": 0.7494387067804221, | |
| "recall": 0.8246047430830039, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8859948068808828, | |
| "precision": 0.8794297680412371, | |
| "recall": 0.8926586003924133, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.8957100204284741, | |
| "precision": 0.9270302504608046, | |
| "recall": 0.8664369679773004, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8338446853480602, | |
| "eval_loss": 0.9773775935173035, | |
| "eval_macro avg": { | |
| "f1-score": 0.751513101933445, | |
| "precision": 0.733920491823513, | |
| "recall": 0.7719563015151424, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0915, | |
| "eval_samples_per_second": 13.133, | |
| "eval_steps_per_second": 1.642, | |
| "eval_weighted avg": { | |
| "f1-score": 0.835464012013009, | |
| "precision": 0.8382307794620859, | |
| "recall": 0.8338446853480602, | |
| "support": 29334.0 | |
| }, | |
| "step": 1134 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5882352941176471, | |
| "precision": 0.5487804878048781, | |
| "recall": 0.6338028169014085, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7338129496402878, | |
| "precision": 0.7445255474452555, | |
| "recall": 0.723404255319149, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7802197802197803, | |
| "precision": 0.7593582887700535, | |
| "recall": 0.8022598870056498, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.6005580097651709, | |
| "precision": 0.570828729281768, | |
| "recall": 0.6335540838852097, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7582104990949057, | |
| "precision": 0.7954422137818774, | |
| "recall": 0.724308300395257, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.885193918204732, | |
| "precision": 0.8803978651140223, | |
| "recall": 0.8900425114453892, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.8958712022851208, | |
| "precision": 0.9188239054010866, | |
| "recall": 0.874037292257803, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8321742687666189, | |
| "eval_loss": 0.9719740152359009, | |
| "eval_macro avg": { | |
| "f1-score": 0.748871664761092, | |
| "precision": 0.7454510053712774, | |
| "recall": 0.7544870210299808, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0974, | |
| "eval_samples_per_second": 13.12, | |
| "eval_steps_per_second": 1.64, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8343275428320325, | |
| "precision": 0.8376519459918353, | |
| "recall": 0.8321742687666189, | |
| "support": 29334.0 | |
| }, | |
| "step": 1215 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5836177474402731, | |
| "precision": 0.5662251655629139, | |
| "recall": 0.602112676056338, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.726643598615917, | |
| "precision": 0.7094594594594594, | |
| "recall": 0.7446808510638298, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7870879120879121, | |
| "precision": 0.766042780748663, | |
| "recall": 0.809322033898305, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5983325159391859, | |
| "precision": 0.5981858298602599, | |
| "recall": 0.5984792739759627, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7764526769855367, | |
| "precision": 0.7981220657276995, | |
| "recall": 0.7559288537549407, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.885249868914613, | |
| "precision": 0.8736565560066873, | |
| "recall": 0.8971550032701112, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9045096930117758, | |
| "precision": 0.9181542958555173, | |
| "recall": 0.8912646939602756, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8382082225403968, | |
| "eval_loss": 1.0037230253219604, | |
| "eval_macro avg": { | |
| "f1-score": 0.751699144713602, | |
| "precision": 0.7471208790316002, | |
| "recall": 0.7569919122828231, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.1157, | |
| "eval_samples_per_second": 13.081, | |
| "eval_steps_per_second": 1.635, | |
| "eval_weighted avg": { | |
| "f1-score": 0.838292846606077, | |
| "precision": 0.8387644471781172, | |
| "recall": 0.8382082225403968, | |
| "support": 29334.0 | |
| }, | |
| "step": 1296 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5652173913043479, | |
| "precision": 0.5382165605095541, | |
| "recall": 0.5950704225352113, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7163120567375887, | |
| "precision": 0.7163120567375887, | |
| "recall": 0.7163120567375887, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7868190988567586, | |
| "precision": 0.7509627727856226, | |
| "recall": 0.826271186440678, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5757906215921483, | |
| "precision": 0.5689655172413793, | |
| "recall": 0.5827814569536424, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7535337959393473, | |
| "precision": 0.7852169255490091, | |
| "recall": 0.724308300395257, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8836841272353221, | |
| "precision": 0.8561790861698866, | |
| "recall": 0.9130150425114454, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.8950797097611355, | |
| "precision": 0.9375346721402419, | |
| "recall": 0.8563032022699635, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8289016158723665, | |
| "eval_loss": 1.0844500064849854, | |
| "eval_macro avg": { | |
| "f1-score": 0.7394909716323783, | |
| "precision": 0.7361982273047546, | |
| "recall": 0.7448659525491124, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.1433, | |
| "eval_samples_per_second": 13.022, | |
| "eval_steps_per_second": 1.628, | |
| "eval_weighted avg": { | |
| "f1-score": 0.829519030769714, | |
| "precision": 0.8324422630439487, | |
| "recall": 0.8289016158723665, | |
| "support": 29334.0 | |
| }, | |
| "step": 1377 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5774647887323944, | |
| "precision": 0.5774647887323944, | |
| "recall": 0.5774647887323944, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7544483985765125, | |
| "precision": 0.7571428571428571, | |
| "recall": 0.75177304964539, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7870182555780934, | |
| "precision": 0.754863813229572, | |
| "recall": 0.8220338983050848, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5765079365079365, | |
| "precision": 0.59768299104792, | |
| "recall": 0.5567819475104243, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.767060030785018, | |
| "precision": 0.7977588046958378, | |
| "recall": 0.7386363636363636, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8867626126838526, | |
| "precision": 0.8586523736600307, | |
| "recall": 0.9167756703727927, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9008560311284047, | |
| "precision": 0.9229297331774211, | |
| "recall": 0.879813538710985, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8356514624667621, | |
| "eval_loss": 1.0617780685424805, | |
| "eval_macro avg": { | |
| "f1-score": 0.7500168648560301, | |
| "precision": 0.7523564802408619, | |
| "recall": 0.7490398938447764, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.2691, | |
| "eval_samples_per_second": 12.761, | |
| "eval_steps_per_second": 1.595, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8340855697185618, | |
| "precision": 0.8340875618543231, | |
| "recall": 0.8356514624667621, | |
| "support": 29334.0 | |
| }, | |
| "step": 1458 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "grad_norm": 1.3275765180587769, | |
| "learning_rate": 1.4814814814814815e-06, | |
| "loss": 0.0228, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5734265734265734, | |
| "precision": 0.5694444444444444, | |
| "recall": 0.5774647887323944, | |
| "support": 284.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7420494699646644, | |
| "precision": 0.7394366197183099, | |
| "recall": 0.7446808510638298, | |
| "support": 141.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7859078590785908, | |
| "precision": 0.7552083333333334, | |
| "recall": 0.8192090395480226, | |
| "support": 708.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5766902119071644, | |
| "precision": 0.5936120488184887, | |
| "recall": 0.5607064017660044, | |
| "support": 4077.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7594099694811801, | |
| "precision": 0.7824947589098532, | |
| "recall": 0.7376482213438735, | |
| "support": 2024.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8857539132157718, | |
| "precision": 0.8594939629316312, | |
| "recall": 0.9136690647482014, | |
| "support": 12232.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9014655441222326, | |
| "precision": 0.9252186899935994, | |
| "recall": 0.8789014997973247, | |
| "support": 9868.0 | |
| }, | |
| "eval_accuracy": 0.8344242176314175, | |
| "eval_loss": 1.0645118951797485, | |
| "eval_macro avg": { | |
| "f1-score": 0.7463862201708825, | |
| "precision": 0.7464155511642371, | |
| "recall": 0.7474685524285215, | |
| "support": 29334.0 | |
| }, | |
| "eval_runtime": 6.0813, | |
| "eval_samples_per_second": 13.155, | |
| "eval_steps_per_second": 1.644, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8332419893084726, | |
| "precision": 0.8334350647066741, | |
| "recall": 0.8344242176314175, | |
| "support": 29334.0 | |
| }, | |
| "step": 1539 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1620, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 2732025441642000.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |