| { | |
| "best_metric": 0.6518611919550361, | |
| "best_model_checkpoint": "drive/MyDrive/NLP_HRAF//Models/HRAF_MultiLabel_Hierarchical/Model_1_BaseTest/Hierarchy_test_fold_1/checkpoint-10790", | |
| "epoch": 13.0, | |
| "eval_steps": 500, | |
| "global_step": 10790, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12048192771084337, | |
| "grad_norm": 4.097501277923584, | |
| "learning_rate": 1.9839357429718877e-05, | |
| "loss": 2.1939, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24096385542168675, | |
| "grad_norm": 3.734555721282959, | |
| "learning_rate": 1.967871485943775e-05, | |
| "loss": 1.7255, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3614457831325301, | |
| "grad_norm": 3.1774344444274902, | |
| "learning_rate": 1.951807228915663e-05, | |
| "loss": 1.5907, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4819277108433735, | |
| "grad_norm": 5.335158824920654, | |
| "learning_rate": 1.9357429718875505e-05, | |
| "loss": 1.5148, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6024096385542169, | |
| "grad_norm": 10.060626983642578, | |
| "learning_rate": 1.9196787148594377e-05, | |
| "loss": 1.4449, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7228915662650602, | |
| "grad_norm": 6.7067084312438965, | |
| "learning_rate": 1.9036144578313255e-05, | |
| "loss": 1.3943, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8433734939759037, | |
| "grad_norm": 5.713496685028076, | |
| "learning_rate": 1.887550200803213e-05, | |
| "loss": 1.3414, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.963855421686747, | |
| "grad_norm": 5.983292102813721, | |
| "learning_rate": 1.8714859437751005e-05, | |
| "loss": 1.3267, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.568414707655214, | |
| "f1": 0.2052586938083121, | |
| "loss": 0.28379881381988525, | |
| "roc_auc": 0.557318161535029 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.5563592525617842, | |
| "f1": 0.12061206120612061, | |
| "loss": 0.4551348090171814, | |
| "roc_auc": 0.5310209707194632 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6865581675708258, | |
| "f1": 0.6954503249767873, | |
| "loss": 0.4480176568031311, | |
| "roc_auc": 0.7856260330258503 | |
| }, | |
| "eval_f1": 0.4580407659608961, | |
| "eval_loss": 1.2922335863113403, | |
| "eval_main_metrics": { | |
| "accuracy": 0.5792646172393008, | |
| "f1": 0.8108419838523645, | |
| "loss": 1.066821813583374, | |
| "roc_auc": 0.8017363757036454 | |
| }, | |
| "eval_runtime": 19.3038, | |
| "eval_samples_per_second": 85.942, | |
| "eval_steps_per_second": 10.775, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.0843373493975903, | |
| "grad_norm": 2.3373818397521973, | |
| "learning_rate": 1.855421686746988e-05, | |
| "loss": 1.2549, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2048192771084336, | |
| "grad_norm": 4.526793956756592, | |
| "learning_rate": 1.8393574297188755e-05, | |
| "loss": 1.184, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3253012048192772, | |
| "grad_norm": 4.537077903747559, | |
| "learning_rate": 1.8232931726907634e-05, | |
| "loss": 1.16, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.4457831325301205, | |
| "grad_norm": 9.092710494995117, | |
| "learning_rate": 1.807228915662651e-05, | |
| "loss": 1.1414, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.5662650602409638, | |
| "grad_norm": 11.58354377746582, | |
| "learning_rate": 1.7911646586345384e-05, | |
| "loss": 1.1569, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.6867469879518073, | |
| "grad_norm": 8.659878730773926, | |
| "learning_rate": 1.775100401606426e-05, | |
| "loss": 1.0936, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.8072289156626506, | |
| "grad_norm": 4.855623245239258, | |
| "learning_rate": 1.7590361445783134e-05, | |
| "loss": 1.1401, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.927710843373494, | |
| "grad_norm": 5.588471412658691, | |
| "learning_rate": 1.742971887550201e-05, | |
| "loss": 1.0588, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.6118143459915611, | |
| "f1": 0.37830319888734354, | |
| "loss": 0.3113616406917572, | |
| "roc_auc": 0.6275566914121131 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.5750452079566004, | |
| "f1": 0.20998278829604128, | |
| "loss": 0.46545448899269104, | |
| "roc_auc": 0.5587718371637969 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6913803496081977, | |
| "f1": 0.7058823529411764, | |
| "loss": 0.5396388173103333, | |
| "roc_auc": 0.7942354115176712 | |
| }, | |
| "eval_f1": 0.5340634299499163, | |
| "eval_loss": 1.181040644645691, | |
| "eval_main_metrics": { | |
| "accuracy": 0.6377335744424352, | |
| "f1": 0.8420853796751038, | |
| "loss": 1.2089903354644775, | |
| "roc_auc": 0.8308065869363559 | |
| }, | |
| "eval_runtime": 19.4619, | |
| "eval_samples_per_second": 85.243, | |
| "eval_steps_per_second": 10.688, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 2.0481927710843375, | |
| "grad_norm": 4.3426384925842285, | |
| "learning_rate": 1.7269076305220884e-05, | |
| "loss": 1.0881, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.1686746987951806, | |
| "grad_norm": 15.190299987792969, | |
| "learning_rate": 1.710843373493976e-05, | |
| "loss": 0.9603, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.289156626506024, | |
| "grad_norm": 10.794776916503906, | |
| "learning_rate": 1.6947791164658637e-05, | |
| "loss": 0.9664, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.4096385542168672, | |
| "grad_norm": 10.562026977539062, | |
| "learning_rate": 1.6787148594377512e-05, | |
| "loss": 0.9303, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.5301204819277108, | |
| "grad_norm": 4.81545877456665, | |
| "learning_rate": 1.6626506024096387e-05, | |
| "loss": 0.9407, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.6506024096385543, | |
| "grad_norm": 4.499901294708252, | |
| "learning_rate": 1.6465863453815262e-05, | |
| "loss": 0.9614, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.7710843373493974, | |
| "grad_norm": 4.810569763183594, | |
| "learning_rate": 1.6305220883534137e-05, | |
| "loss": 0.9482, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.891566265060241, | |
| "grad_norm": 8.373404502868652, | |
| "learning_rate": 1.6144578313253015e-05, | |
| "loss": 0.8748, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.621458710066305, | |
| "f1": 0.4234990316333118, | |
| "loss": 0.31134167313575745, | |
| "roc_auc": 0.6527716928319338 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.5822784810126582, | |
| "f1": 0.2585924713584288, | |
| "loss": 0.489061564207077, | |
| "roc_auc": 0.5754498216809775 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6937914406268837, | |
| "f1": 0.7111111111111111, | |
| "loss": 0.6078017354011536, | |
| "roc_auc": 0.798155101571195 | |
| }, | |
| "eval_f1": 0.5597986447252539, | |
| "eval_loss": 1.1940369606018066, | |
| "eval_main_metrics": { | |
| "accuracy": 0.6497890295358649, | |
| "f1": 0.8459919647981634, | |
| "loss": 1.356093406677246, | |
| "roc_auc": 0.8375618095347757 | |
| }, | |
| "eval_runtime": 19.4292, | |
| "eval_samples_per_second": 85.387, | |
| "eval_steps_per_second": 10.706, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 3.0120481927710845, | |
| "grad_norm": 7.537188529968262, | |
| "learning_rate": 1.5983935742971887e-05, | |
| "loss": 0.9201, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.1325301204819276, | |
| "grad_norm": 6.753769397735596, | |
| "learning_rate": 1.5823293172690762e-05, | |
| "loss": 0.7751, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.253012048192771, | |
| "grad_norm": 6.366585731506348, | |
| "learning_rate": 1.566265060240964e-05, | |
| "loss": 0.7857, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.3734939759036147, | |
| "grad_norm": 8.141886711120605, | |
| "learning_rate": 1.5502008032128516e-05, | |
| "loss": 0.7844, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.4939759036144578, | |
| "grad_norm": 12.003459930419922, | |
| "learning_rate": 1.534136546184739e-05, | |
| "loss": 0.7853, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.6144578313253013, | |
| "grad_norm": 4.899393558502197, | |
| "learning_rate": 1.5180722891566266e-05, | |
| "loss": 0.7652, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.734939759036145, | |
| "grad_norm": 7.926997661590576, | |
| "learning_rate": 1.5020080321285142e-05, | |
| "loss": 0.7908, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 3.855421686746988, | |
| "grad_norm": 3.9363718032836914, | |
| "learning_rate": 1.4859437751004017e-05, | |
| "loss": 0.8142, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.9759036144578315, | |
| "grad_norm": 4.785311222076416, | |
| "learning_rate": 1.4698795180722894e-05, | |
| "loss": 0.75, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.6262808921036769, | |
| "f1": 0.47516457211250746, | |
| "loss": 0.3067108690738678, | |
| "roc_auc": 0.6846639081578841 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.6100060277275468, | |
| "f1": 0.38606271777003487, | |
| "loss": 0.5579449534416199, | |
| "roc_auc": 0.6297609166955901 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6955997588908981, | |
| "f1": 0.7214808437365476, | |
| "loss": 0.6555235981941223, | |
| "roc_auc": 0.8122881121542476 | |
| }, | |
| "eval_f1": 0.6080738826170755, | |
| "eval_loss": 1.2421071529388428, | |
| "eval_main_metrics": { | |
| "accuracy": 0.6594333936106088, | |
| "f1": 0.8495873968492123, | |
| "loss": 1.4700279235839844, | |
| "roc_auc": 0.8373546457094286 | |
| }, | |
| "eval_runtime": 19.3706, | |
| "eval_samples_per_second": 85.645, | |
| "eval_steps_per_second": 10.738, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 4.096385542168675, | |
| "grad_norm": 2.819312334060669, | |
| "learning_rate": 1.4538152610441769e-05, | |
| "loss": 0.6905, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.216867469879518, | |
| "grad_norm": 15.132315635681152, | |
| "learning_rate": 1.4377510040160642e-05, | |
| "loss": 0.66, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.337349397590361, | |
| "grad_norm": 13.252199172973633, | |
| "learning_rate": 1.4216867469879519e-05, | |
| "loss": 0.6854, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.457831325301205, | |
| "grad_norm": 3.5887160301208496, | |
| "learning_rate": 1.4056224899598394e-05, | |
| "loss": 0.6423, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 4.578313253012048, | |
| "grad_norm": 3.583491563796997, | |
| "learning_rate": 1.3895582329317269e-05, | |
| "loss": 0.6372, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 4.698795180722891, | |
| "grad_norm": 9.358288764953613, | |
| "learning_rate": 1.3734939759036146e-05, | |
| "loss": 0.6457, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 4.8192771084337345, | |
| "grad_norm": 6.61803674697876, | |
| "learning_rate": 1.357429718875502e-05, | |
| "loss": 0.6402, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.9397590361445785, | |
| "grad_norm": 12.123319625854492, | |
| "learning_rate": 1.3413654618473897e-05, | |
| "loss": 0.6447, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.6274864376130199, | |
| "f1": 0.47335025380710655, | |
| "loss": 0.4026564955711365, | |
| "roc_auc": 0.6765029816234635 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.6244725738396625, | |
| "f1": 0.42625169147496617, | |
| "loss": 0.5328236222267151, | |
| "roc_auc": 0.6485007620686012 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6949969861362266, | |
| "f1": 0.7206896551724137, | |
| "loss": 0.7433874011039734, | |
| "roc_auc": 0.8115834972530086 | |
| }, | |
| "eval_f1": 0.6173571829119144, | |
| "eval_loss": 1.272179365158081, | |
| "eval_main_metrics": { | |
| "accuracy": 0.650994575045208, | |
| "f1": 0.8491371311931712, | |
| "loss": 1.7594246864318848, | |
| "roc_auc": 0.8347010566163695 | |
| }, | |
| "eval_runtime": 19.3728, | |
| "eval_samples_per_second": 85.635, | |
| "eval_steps_per_second": 10.737, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 5.0602409638554215, | |
| "grad_norm": 4.934209823608398, | |
| "learning_rate": 1.3253012048192772e-05, | |
| "loss": 0.5871, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.180722891566265, | |
| "grad_norm": 3.2406387329101562, | |
| "learning_rate": 1.309236947791165e-05, | |
| "loss": 0.5427, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 5.301204819277109, | |
| "grad_norm": 5.232382774353027, | |
| "learning_rate": 1.2931726907630524e-05, | |
| "loss": 0.5588, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.421686746987952, | |
| "grad_norm": 14.17167854309082, | |
| "learning_rate": 1.2771084337349398e-05, | |
| "loss": 0.5816, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.542168674698795, | |
| "grad_norm": 15.979377746582031, | |
| "learning_rate": 1.2610441767068273e-05, | |
| "loss": 0.5946, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 5.662650602409639, | |
| "grad_norm": 7.11401891708374, | |
| "learning_rate": 1.244979919678715e-05, | |
| "loss": 0.5395, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 5.783132530120482, | |
| "grad_norm": 2.767531156539917, | |
| "learning_rate": 1.2289156626506024e-05, | |
| "loss": 0.5364, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.903614457831325, | |
| "grad_norm": 4.462075233459473, | |
| "learning_rate": 1.2128514056224901e-05, | |
| "loss": 0.5586, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.6359252561784208, | |
| "f1": 0.4963144963144963, | |
| "loss": 0.3993087410926819, | |
| "roc_auc": 0.6909881952050626 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.6172393007836046, | |
| "f1": 0.4432786885245902, | |
| "loss": 0.5477549433708191, | |
| "roc_auc": 0.6586722400792753 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6895720313441832, | |
| "f1": 0.7254658385093167, | |
| "loss": 0.775439441204071, | |
| "roc_auc": 0.8210486138922756 | |
| }, | |
| "eval_f1": 0.628022957090326, | |
| "eval_loss": 1.3330581188201904, | |
| "eval_main_metrics": { | |
| "accuracy": 0.6377335744424352, | |
| "f1": 0.8470328050129008, | |
| "loss": 1.7559667825698853, | |
| "roc_auc": 0.8309945868949555 | |
| }, | |
| "eval_runtime": 19.4268, | |
| "eval_samples_per_second": 85.398, | |
| "eval_steps_per_second": 10.707, | |
| "step": 4980 | |
| }, | |
| { | |
| "epoch": 6.024096385542169, | |
| "grad_norm": 4.994393825531006, | |
| "learning_rate": 1.1967871485943776e-05, | |
| "loss": 0.5326, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.144578313253012, | |
| "grad_norm": 2.5982210636138916, | |
| "learning_rate": 1.1807228915662651e-05, | |
| "loss": 0.4935, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 6.265060240963855, | |
| "grad_norm": 9.731000900268555, | |
| "learning_rate": 1.1646586345381528e-05, | |
| "loss": 0.4444, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 6.385542168674699, | |
| "grad_norm": 3.476001501083374, | |
| "learning_rate": 1.1485943775100403e-05, | |
| "loss": 0.4772, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 6.506024096385542, | |
| "grad_norm": 9.325688362121582, | |
| "learning_rate": 1.132530120481928e-05, | |
| "loss": 0.4758, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 6.626506024096385, | |
| "grad_norm": 10.23697566986084, | |
| "learning_rate": 1.1164658634538153e-05, | |
| "loss": 0.467, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 6.746987951807229, | |
| "grad_norm": 6.919719219207764, | |
| "learning_rate": 1.1004016064257028e-05, | |
| "loss": 0.4596, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 6.867469879518072, | |
| "grad_norm": 3.161492347717285, | |
| "learning_rate": 1.0843373493975904e-05, | |
| "loss": 0.4901, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 6.9879518072289155, | |
| "grad_norm": 5.633391857147217, | |
| "learning_rate": 1.068273092369478e-05, | |
| "loss": 0.5053, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.6347197106690777, | |
| "f1": 0.4936014625228519, | |
| "loss": 0.44190824031829834, | |
| "roc_auc": 0.6908239016672751 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.6208559373116335, | |
| "f1": 0.4570337364735837, | |
| "loss": 0.5815909504890442, | |
| "roc_auc": 0.6677868823597467 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6714888487040386, | |
| "f1": 0.7207713941341904, | |
| "loss": 0.7780100703239441, | |
| "roc_auc": 0.8228283501014495 | |
| }, | |
| "eval_f1": 0.6299279587105135, | |
| "eval_loss": 1.403409481048584, | |
| "eval_main_metrics": { | |
| "accuracy": 0.6443640747438216, | |
| "f1": 0.8483052417114281, | |
| "loss": 2.0327484607696533, | |
| "roc_auc": 0.8334165115229015 | |
| }, | |
| "eval_runtime": 19.4509, | |
| "eval_samples_per_second": 85.292, | |
| "eval_steps_per_second": 10.694, | |
| "step": 5810 | |
| }, | |
| { | |
| "epoch": 7.108433734939759, | |
| "grad_norm": 2.695713996887207, | |
| "learning_rate": 1.0522088353413654e-05, | |
| "loss": 0.4098, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 7.228915662650603, | |
| "grad_norm": 3.300992727279663, | |
| "learning_rate": 1.0361445783132531e-05, | |
| "loss": 0.4406, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.349397590361446, | |
| "grad_norm": 9.012105941772461, | |
| "learning_rate": 1.0200803212851406e-05, | |
| "loss": 0.3848, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 7.469879518072289, | |
| "grad_norm": 1.8880501985549927, | |
| "learning_rate": 1.0040160642570283e-05, | |
| "loss": 0.4086, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 7.590361445783133, | |
| "grad_norm": 6.892731666564941, | |
| "learning_rate": 9.879518072289156e-06, | |
| "loss": 0.4284, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 7.710843373493976, | |
| "grad_norm": 7.901873588562012, | |
| "learning_rate": 9.718875502008033e-06, | |
| "loss": 0.45, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 7.831325301204819, | |
| "grad_norm": 4.076874256134033, | |
| "learning_rate": 9.558232931726908e-06, | |
| "loss": 0.4319, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 7.951807228915663, | |
| "grad_norm": 4.073435306549072, | |
| "learning_rate": 9.397590361445785e-06, | |
| "loss": 0.42, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.6395418927064497, | |
| "f1": 0.5042735042735043, | |
| "loss": 0.45456641912460327, | |
| "roc_auc": 0.6954778710802808 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.6298975286317059, | |
| "f1": 0.4623179227359088, | |
| "loss": 0.5961005687713623, | |
| "roc_auc": 0.6706782159043465 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6853526220614828, | |
| "f1": 0.7223140495867769, | |
| "loss": 0.8032154440879822, | |
| "roc_auc": 0.8192863575314914 | |
| }, | |
| "eval_f1": 0.6344324161815224, | |
| "eval_loss": 1.4469618797302246, | |
| "eval_main_metrics": { | |
| "accuracy": 0.646172393007836, | |
| "f1": 0.8488241881298993, | |
| "loss": 2.039750337600708, | |
| "roc_auc": 0.835544105032868 | |
| }, | |
| "eval_runtime": 19.422, | |
| "eval_samples_per_second": 85.419, | |
| "eval_steps_per_second": 10.71, | |
| "step": 6640 | |
| }, | |
| { | |
| "epoch": 8.072289156626505, | |
| "grad_norm": 2.453636407852173, | |
| "learning_rate": 9.23694779116466e-06, | |
| "loss": 0.3554, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 8.19277108433735, | |
| "grad_norm": 3.2704885005950928, | |
| "learning_rate": 9.076305220883535e-06, | |
| "loss": 0.3916, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 8.313253012048193, | |
| "grad_norm": 5.025237083435059, | |
| "learning_rate": 8.91566265060241e-06, | |
| "loss": 0.3619, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 8.433734939759036, | |
| "grad_norm": 7.339418888092041, | |
| "learning_rate": 8.755020080321286e-06, | |
| "loss": 0.345, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.55421686746988, | |
| "grad_norm": 11.077627182006836, | |
| "learning_rate": 8.594377510040161e-06, | |
| "loss": 0.3704, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 8.674698795180722, | |
| "grad_norm": 4.094336986541748, | |
| "learning_rate": 8.433734939759038e-06, | |
| "loss": 0.4081, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 8.795180722891565, | |
| "grad_norm": 2.9691665172576904, | |
| "learning_rate": 8.273092369477911e-06, | |
| "loss": 0.3692, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 8.91566265060241, | |
| "grad_norm": 2.072866916656494, | |
| "learning_rate": 8.112449799196788e-06, | |
| "loss": 0.3552, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.6305003013863774, | |
| "f1": 0.509478672985782, | |
| "loss": 0.4554847180843353, | |
| "roc_auc": 0.7022230335483347 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.6106088004822182, | |
| "f1": 0.43440594059405946, | |
| "loss": 0.6525366902351379, | |
| "roc_auc": 0.6608236377080599 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6799276672694394, | |
| "f1": 0.724025974025974, | |
| "loss": 0.9302711486816406, | |
| "roc_auc": 0.8233804035411074 | |
| }, | |
| "eval_f1": 0.6287379744042451, | |
| "eval_loss": 1.5192142724990845, | |
| "eval_main_metrics": { | |
| "accuracy": 0.6449668474984931, | |
| "f1": 0.8470413100111649, | |
| "loss": 2.102555513381958, | |
| "roc_auc": 0.8330059034412807 | |
| }, | |
| "eval_runtime": 19.4322, | |
| "eval_samples_per_second": 85.374, | |
| "eval_steps_per_second": 10.704, | |
| "step": 7470 | |
| }, | |
| { | |
| "epoch": 9.036144578313253, | |
| "grad_norm": 6.187836170196533, | |
| "learning_rate": 7.951807228915663e-06, | |
| "loss": 0.3559, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.156626506024097, | |
| "grad_norm": 9.382346153259277, | |
| "learning_rate": 7.79116465863454e-06, | |
| "loss": 0.3518, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 9.27710843373494, | |
| "grad_norm": 3.6472980976104736, | |
| "learning_rate": 7.630522088353415e-06, | |
| "loss": 0.3199, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 9.397590361445783, | |
| "grad_norm": 6.128402233123779, | |
| "learning_rate": 7.469879518072289e-06, | |
| "loss": 0.3212, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 9.518072289156626, | |
| "grad_norm": 2.361755132675171, | |
| "learning_rate": 7.309236947791165e-06, | |
| "loss": 0.3245, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 9.638554216867469, | |
| "grad_norm": 4.379687786102295, | |
| "learning_rate": 7.148594377510041e-06, | |
| "loss": 0.3294, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 9.759036144578314, | |
| "grad_norm": 6.3661370277404785, | |
| "learning_rate": 6.987951807228917e-06, | |
| "loss": 0.3325, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 9.879518072289157, | |
| "grad_norm": 2.1560158729553223, | |
| "learning_rate": 6.8273092369477925e-06, | |
| "loss": 0.3192, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.878142356872559, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.3227, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.6317058468957203, | |
| "f1": 0.5257076834199884, | |
| "loss": 0.3950282335281372, | |
| "roc_auc": 0.7138452801103403 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.6190476190476191, | |
| "f1": 0.48321342925659466, | |
| "loss": 0.6035521626472473, | |
| "roc_auc": 0.6868496637340858 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6949969861362266, | |
| "f1": 0.73347022587269, | |
| "loss": 1.0751080513000488, | |
| "roc_auc": 0.8277649690562431 | |
| }, | |
| "eval_f1": 0.6476335229111712, | |
| "eval_loss": 1.5520377159118652, | |
| "eval_main_metrics": { | |
| "accuracy": 0.6353224834237492, | |
| "f1": 0.8481427530954114, | |
| "loss": 2.005683422088623, | |
| "roc_auc": 0.8296800426683087 | |
| }, | |
| "eval_runtime": 19.4155, | |
| "eval_samples_per_second": 85.447, | |
| "eval_steps_per_second": 10.713, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 10.120481927710843, | |
| "grad_norm": 4.528349876403809, | |
| "learning_rate": 6.5060240963855425e-06, | |
| "loss": 0.3223, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 10.240963855421686, | |
| "grad_norm": 6.405284881591797, | |
| "learning_rate": 6.345381526104418e-06, | |
| "loss": 0.2805, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 10.36144578313253, | |
| "grad_norm": 4.246714115142822, | |
| "learning_rate": 6.184738955823294e-06, | |
| "loss": 0.2912, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 10.481927710843374, | |
| "grad_norm": 2.8840444087982178, | |
| "learning_rate": 6.02409638554217e-06, | |
| "loss": 0.303, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 10.602409638554217, | |
| "grad_norm": 6.155231952667236, | |
| "learning_rate": 5.863453815261044e-06, | |
| "loss": 0.289, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 10.72289156626506, | |
| "grad_norm": 5.590630531311035, | |
| "learning_rate": 5.70281124497992e-06, | |
| "loss": 0.2866, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 10.843373493975903, | |
| "grad_norm": 2.1477112770080566, | |
| "learning_rate": 5.542168674698796e-06, | |
| "loss": 0.2908, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 10.963855421686747, | |
| "grad_norm": 3.363445520401001, | |
| "learning_rate": 5.381526104417672e-06, | |
| "loss": 0.3135, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.6365280289330922, | |
| "f1": 0.5067319461444308, | |
| "loss": 0.4702241122722626, | |
| "roc_auc": 0.6962618149365137 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.621458710066305, | |
| "f1": 0.4755501222493888, | |
| "loss": 0.6331600546836853, | |
| "roc_auc": 0.680847681852707 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6943942133815552, | |
| "f1": 0.73487616727568, | |
| "loss": 0.8929408192634583, | |
| "roc_auc": 0.8306706066392446 | |
| }, | |
| "eval_f1": 0.6414051572871415, | |
| "eval_loss": 1.5805182456970215, | |
| "eval_main_metrics": { | |
| "accuracy": 0.6401446654611211, | |
| "f1": 0.8484623934790664, | |
| "loss": 2.2667479515075684, | |
| "roc_auc": 0.8336254542725883 | |
| }, | |
| "eval_runtime": 19.3897, | |
| "eval_samples_per_second": 85.561, | |
| "eval_steps_per_second": 10.727, | |
| "step": 9130 | |
| }, | |
| { | |
| "epoch": 11.08433734939759, | |
| "grad_norm": 3.677710771560669, | |
| "learning_rate": 5.220883534136547e-06, | |
| "loss": 0.2626, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 11.204819277108435, | |
| "grad_norm": 3.71104097366333, | |
| "learning_rate": 5.060240963855422e-06, | |
| "loss": 0.2811, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 11.325301204819278, | |
| "grad_norm": 4.658101558685303, | |
| "learning_rate": 4.899598393574298e-06, | |
| "loss": 0.2556, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 11.44578313253012, | |
| "grad_norm": 2.3544187545776367, | |
| "learning_rate": 4.7389558232931736e-06, | |
| "loss": 0.2675, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 11.566265060240964, | |
| "grad_norm": 2.9858105182647705, | |
| "learning_rate": 4.578313253012049e-06, | |
| "loss": 0.2601, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 11.686746987951807, | |
| "grad_norm": 5.517003059387207, | |
| "learning_rate": 4.4176706827309244e-06, | |
| "loss": 0.2706, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 11.80722891566265, | |
| "grad_norm": 5.479954242706299, | |
| "learning_rate": 4.2570281124497995e-06, | |
| "loss": 0.2678, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 11.927710843373493, | |
| "grad_norm": 3.5991973876953125, | |
| "learning_rate": 4.096385542168675e-06, | |
| "loss": 0.2555, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.6329113924050633, | |
| "f1": 0.5242030696576152, | |
| "loss": 0.41705745458602905, | |
| "roc_auc": 0.7097399699809339 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.6262808921036769, | |
| "f1": 0.5017221584385763, | |
| "loss": 0.6140196919441223, | |
| "roc_auc": 0.7016332915830403 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6871609403254972, | |
| "f1": 0.7303643724696356, | |
| "loss": 1.0149368047714233, | |
| "roc_auc": 0.8280933246892902 | |
| }, | |
| "eval_f1": 0.6511378774267244, | |
| "eval_loss": 1.6208207607269287, | |
| "eval_main_metrics": { | |
| "accuracy": 0.6395418927064497, | |
| "f1": 0.8482619091410705, | |
| "loss": 2.2304847240448, | |
| "roc_auc": 0.8319157462619947 | |
| }, | |
| "eval_runtime": 19.4554, | |
| "eval_samples_per_second": 85.272, | |
| "eval_steps_per_second": 10.691, | |
| "step": 9960 | |
| }, | |
| { | |
| "epoch": 12.048192771084338, | |
| "grad_norm": 4.944530010223389, | |
| "learning_rate": 3.93574297188755e-06, | |
| "loss": 0.2796, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.168674698795181, | |
| "grad_norm": 4.537522792816162, | |
| "learning_rate": 3.7751004016064258e-06, | |
| "loss": 0.2397, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 12.289156626506024, | |
| "grad_norm": 5.054214954376221, | |
| "learning_rate": 3.6144578313253016e-06, | |
| "loss": 0.2311, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 12.409638554216867, | |
| "grad_norm": 1.6431657075881958, | |
| "learning_rate": 3.453815261044177e-06, | |
| "loss": 0.2515, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 12.53012048192771, | |
| "grad_norm": 2.898925304412842, | |
| "learning_rate": 3.2931726907630525e-06, | |
| "loss": 0.2424, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 12.650602409638553, | |
| "grad_norm": 3.5978894233703613, | |
| "learning_rate": 3.132530120481928e-06, | |
| "loss": 0.2573, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 12.771084337349398, | |
| "grad_norm": 3.019956588745117, | |
| "learning_rate": 2.9718875502008034e-06, | |
| "loss": 0.2469, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 12.891566265060241, | |
| "grad_norm": 4.129029273986816, | |
| "learning_rate": 2.811244979919679e-06, | |
| "loss": 0.2377, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_action_metrics": { | |
| "accuracy": 0.6298975286317059, | |
| "f1": 0.5311438278595696, | |
| "loss": 0.40619078278541565, | |
| "roc_auc": 0.7197446351060808 | |
| }, | |
| "eval_cause_metrics": { | |
| "accuracy": 0.6220614828209765, | |
| "f1": 0.5011520737327189, | |
| "loss": 0.6362375617027283, | |
| "roc_auc": 0.7008556294988456 | |
| }, | |
| "eval_event_metrics": { | |
| "accuracy": 0.6841470765521398, | |
| "f1": 0.729192042224929, | |
| "loss": 1.0542970895767212, | |
| "roc_auc": 0.8268163000692555 | |
| }, | |
| "eval_f1": 0.6518611919550361, | |
| "eval_loss": 1.6575490236282349, | |
| "eval_main_metrics": { | |
| "accuracy": 0.6305003013863774, | |
| "f1": 0.8459568240029272, | |
| "loss": 2.175412654876709, | |
| "roc_auc": 0.8282710537312776 | |
| }, | |
| "eval_runtime": 19.3864, | |
| "eval_samples_per_second": 85.575, | |
| "eval_steps_per_second": 10.729, | |
| "step": 10790 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 12450, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7890557846663520.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |