| { | |
| "best_metric": 0.8164585392148055, | |
| "best_model_checkpoint": "./XLM-V-multi-outputs/checkpoint-40000", | |
| "epoch": 19.82300884955752, | |
| "eval_steps": 1000, | |
| "global_step": 56000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.35398230088495575, | |
| "grad_norm": 1.4846587181091309, | |
| "learning_rate": 3.5398230088495575e-07, | |
| "loss": 0.6947, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.35398230088495575, | |
| "eval_accuracy": 0.5090112516180424, | |
| "eval_f1": 0.6746288353678654, | |
| "eval_loss": 0.6931052207946777, | |
| "eval_precision": 0.5090112516180424, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 53.8213, | |
| "eval_samples_per_second": 186.599, | |
| "eval_steps_per_second": 5.834, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7079646017699115, | |
| "grad_norm": 3.2122230529785156, | |
| "learning_rate": 7.079646017699115e-07, | |
| "loss": 0.6924, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7079646017699115, | |
| "eval_accuracy": 0.5752265259384646, | |
| "eval_f1": 0.698601102161933, | |
| "eval_loss": 0.681232213973999, | |
| "eval_precision": 0.5467816854678168, | |
| "eval_recall": 0.9671361502347418, | |
| "eval_runtime": 53.8929, | |
| "eval_samples_per_second": 186.351, | |
| "eval_steps_per_second": 5.826, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0619469026548674, | |
| "grad_norm": 4.78421688079834, | |
| "learning_rate": 1.0619469026548673e-06, | |
| "loss": 0.6783, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0619469026548674, | |
| "eval_accuracy": 0.6099770984765508, | |
| "eval_f1": 0.7189092213850018, | |
| "eval_loss": 0.6526923179626465, | |
| "eval_precision": 0.5677207299104613, | |
| "eval_recall": 0.9798513302034428, | |
| "eval_runtime": 53.7592, | |
| "eval_samples_per_second": 186.814, | |
| "eval_steps_per_second": 5.841, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.415929203539823, | |
| "grad_norm": 49.178401947021484, | |
| "learning_rate": 1.415929203539823e-06, | |
| "loss": 0.6613, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.415929203539823, | |
| "eval_accuracy": 0.6256098775266354, | |
| "eval_f1": 0.7167821633021995, | |
| "eval_loss": 0.633762538433075, | |
| "eval_precision": 0.5828025477707006, | |
| "eval_recall": 0.9307511737089202, | |
| "eval_runtime": 53.5833, | |
| "eval_samples_per_second": 187.428, | |
| "eval_steps_per_second": 5.86, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.7699115044247788, | |
| "grad_norm": 11.674603462219238, | |
| "learning_rate": 1.7699115044247788e-06, | |
| "loss": 0.6551, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.7699115044247788, | |
| "eval_accuracy": 0.6342726277008862, | |
| "eval_f1": 0.7293493478741434, | |
| "eval_loss": 0.6212999820709229, | |
| "eval_precision": 0.5850573353824329, | |
| "eval_recall": 0.968114241001565, | |
| "eval_runtime": 53.825, | |
| "eval_samples_per_second": 186.586, | |
| "eval_steps_per_second": 5.834, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.1238938053097347, | |
| "grad_norm": 5.5286545753479, | |
| "learning_rate": 1.986234021632252e-06, | |
| "loss": 0.6481, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.1238938053097347, | |
| "eval_accuracy": 0.656078860898138, | |
| "eval_f1": 0.740222623345367, | |
| "eval_loss": 0.6046485900878906, | |
| "eval_precision": 0.6012952101661779, | |
| "eval_recall": 0.9626369327073553, | |
| "eval_runtime": 53.8663, | |
| "eval_samples_per_second": 186.443, | |
| "eval_steps_per_second": 5.829, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.47787610619469, | |
| "grad_norm": 47.20358657836914, | |
| "learning_rate": 1.9469026548672563e-06, | |
| "loss": 0.639, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.47787610619469, | |
| "eval_accuracy": 0.6773872348899731, | |
| "eval_f1": 0.7519142419601837, | |
| "eval_loss": 0.5804159045219421, | |
| "eval_precision": 0.6177654755913438, | |
| "eval_recall": 0.9604851330203443, | |
| "eval_runtime": 53.8098, | |
| "eval_samples_per_second": 186.639, | |
| "eval_steps_per_second": 5.835, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.831858407079646, | |
| "grad_norm": 20.222881317138672, | |
| "learning_rate": 1.9075712881022612e-06, | |
| "loss": 0.63, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.831858407079646, | |
| "eval_accuracy": 0.6816688240565568, | |
| "eval_f1": 0.6923299008757579, | |
| "eval_loss": 0.584140419960022, | |
| "eval_precision": 0.6813790490623224, | |
| "eval_recall": 0.7036384976525821, | |
| "eval_runtime": 53.9333, | |
| "eval_samples_per_second": 186.211, | |
| "eval_steps_per_second": 5.822, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.185840707964602, | |
| "grad_norm": 8.778471946716309, | |
| "learning_rate": 1.8682399213372663e-06, | |
| "loss": 0.6087, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.185840707964602, | |
| "eval_accuracy": 0.7025789106840585, | |
| "eval_f1": 0.747271342753194, | |
| "eval_loss": 0.5566529035568237, | |
| "eval_precision": 0.6584165796928582, | |
| "eval_recall": 0.863849765258216, | |
| "eval_runtime": 53.826, | |
| "eval_samples_per_second": 186.583, | |
| "eval_steps_per_second": 5.834, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.5398230088495577, | |
| "grad_norm": 16.00111198425293, | |
| "learning_rate": 1.8289085545722714e-06, | |
| "loss": 0.5851, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.5398230088495577, | |
| "eval_accuracy": 0.708553221149059, | |
| "eval_f1": 0.7571155920670484, | |
| "eval_loss": 0.5440040826797485, | |
| "eval_precision": 0.6574434356535523, | |
| "eval_recall": 0.8924100156494522, | |
| "eval_runtime": 53.8534, | |
| "eval_samples_per_second": 186.488, | |
| "eval_steps_per_second": 5.831, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.893805309734513, | |
| "grad_norm": 4.33933162689209, | |
| "learning_rate": 1.7895771878072763e-06, | |
| "loss": 0.5516, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.893805309734513, | |
| "eval_accuracy": 0.7223937070596436, | |
| "eval_f1": 0.7767456758488148, | |
| "eval_loss": 0.4870859384536743, | |
| "eval_precision": 0.6575379609544468, | |
| "eval_recall": 0.9487480438184663, | |
| "eval_runtime": 53.9228, | |
| "eval_samples_per_second": 186.248, | |
| "eval_steps_per_second": 5.823, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.247787610619469, | |
| "grad_norm": 13.392922401428223, | |
| "learning_rate": 1.7502458210422812e-06, | |
| "loss": 0.495, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.247787610619469, | |
| "eval_accuracy": 0.7339440406253112, | |
| "eval_f1": 0.7293903180068868, | |
| "eval_loss": 0.43779730796813965, | |
| "eval_precision": 0.7561948761024779, | |
| "eval_recall": 0.7044209702660407, | |
| "eval_runtime": 53.9151, | |
| "eval_samples_per_second": 186.274, | |
| "eval_steps_per_second": 5.824, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.601769911504425, | |
| "grad_norm": 6.77086067199707, | |
| "learning_rate": 1.7109144542772859e-06, | |
| "loss": 0.4655, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.601769911504425, | |
| "eval_accuracy": 0.7499751070397291, | |
| "eval_f1": 0.7768991559306975, | |
| "eval_loss": 0.41514188051223755, | |
| "eval_precision": 0.7117043789679309, | |
| "eval_recall": 0.8552425665101722, | |
| "eval_runtime": 53.8935, | |
| "eval_samples_per_second": 186.349, | |
| "eval_steps_per_second": 5.826, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.95575221238938, | |
| "grad_norm": 6.903416156768799, | |
| "learning_rate": 1.671583087512291e-06, | |
| "loss": 0.4451, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.95575221238938, | |
| "eval_accuracy": 0.7510703972916459, | |
| "eval_f1": 0.7728511720879521, | |
| "eval_loss": 0.4052245318889618, | |
| "eval_precision": 0.7215812690872073, | |
| "eval_recall": 0.831964006259781, | |
| "eval_runtime": 53.8879, | |
| "eval_samples_per_second": 186.368, | |
| "eval_steps_per_second": 5.827, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 5.3097345132743365, | |
| "grad_norm": 24.238162994384766, | |
| "learning_rate": 1.6322517207472959e-06, | |
| "loss": 0.4275, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 5.3097345132743365, | |
| "eval_accuracy": 0.7568455640744798, | |
| "eval_f1": 0.7681792291627112, | |
| "eval_loss": 0.3967246115207672, | |
| "eval_precision": 0.7462191073404648, | |
| "eval_recall": 0.7914710485133021, | |
| "eval_runtime": 53.89, | |
| "eval_samples_per_second": 186.361, | |
| "eval_steps_per_second": 5.827, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 5.663716814159292, | |
| "grad_norm": 22.419418334960938, | |
| "learning_rate": 1.5929203539823008e-06, | |
| "loss": 0.4085, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.663716814159292, | |
| "eval_accuracy": 0.7561485611868963, | |
| "eval_f1": 0.7775456444727041, | |
| "eval_loss": 0.39186546206474304, | |
| "eval_precision": 0.7257927759877904, | |
| "eval_recall": 0.837245696400626, | |
| "eval_runtime": 54.0231, | |
| "eval_samples_per_second": 185.902, | |
| "eval_steps_per_second": 5.812, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 6.017699115044247, | |
| "grad_norm": 25.710134506225586, | |
| "learning_rate": 1.5535889872173059e-06, | |
| "loss": 0.4088, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 6.017699115044247, | |
| "eval_accuracy": 0.7595340037837299, | |
| "eval_f1": 0.7705027083531313, | |
| "eval_loss": 0.3929796516895294, | |
| "eval_precision": 0.7492145629273702, | |
| "eval_recall": 0.793035993740219, | |
| "eval_runtime": 53.8892, | |
| "eval_samples_per_second": 186.364, | |
| "eval_steps_per_second": 5.827, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 6.371681415929204, | |
| "grad_norm": 12.00070571899414, | |
| "learning_rate": 1.5142576204523108e-06, | |
| "loss": 0.3982, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 6.371681415929204, | |
| "eval_accuracy": 0.756347704869063, | |
| "eval_f1": 0.7932753231393089, | |
| "eval_loss": 0.38643109798431396, | |
| "eval_precision": 0.6981412639405205, | |
| "eval_recall": 0.9184272300469484, | |
| "eval_runtime": 53.924, | |
| "eval_samples_per_second": 186.243, | |
| "eval_steps_per_second": 5.823, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 6.725663716814159, | |
| "grad_norm": 8.227592468261719, | |
| "learning_rate": 1.4749262536873155e-06, | |
| "loss": 0.3952, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 6.725663716814159, | |
| "eval_accuracy": 0.7651100268843971, | |
| "eval_f1": 0.776673293571902, | |
| "eval_loss": 0.3926367461681366, | |
| "eval_precision": 0.7525224729407448, | |
| "eval_recall": 0.8024256651017214, | |
| "eval_runtime": 53.9659, | |
| "eval_samples_per_second": 186.099, | |
| "eval_steps_per_second": 5.818, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 7.079646017699115, | |
| "grad_norm": 5.537351131439209, | |
| "learning_rate": 1.4355948869223204e-06, | |
| "loss": 0.3876, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 7.079646017699115, | |
| "eval_accuracy": 0.763915164791397, | |
| "eval_f1": 0.792472647702407, | |
| "eval_loss": 0.38483595848083496, | |
| "eval_precision": 0.7170917155076826, | |
| "eval_recall": 0.8855633802816901, | |
| "eval_runtime": 53.9366, | |
| "eval_samples_per_second": 186.2, | |
| "eval_steps_per_second": 5.822, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 7.433628318584071, | |
| "grad_norm": 4.749475002288818, | |
| "learning_rate": 1.3962635201573255e-06, | |
| "loss": 0.3832, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 7.433628318584071, | |
| "eval_accuracy": 0.7616250124464802, | |
| "eval_f1": 0.7432432432432432, | |
| "eval_loss": 0.38059765100479126, | |
| "eval_precision": 0.8226495726495726, | |
| "eval_recall": 0.6778169014084507, | |
| "eval_runtime": 53.9079, | |
| "eval_samples_per_second": 186.299, | |
| "eval_steps_per_second": 5.825, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 7.787610619469026, | |
| "grad_norm": 9.643062591552734, | |
| "learning_rate": 1.3569321533923304e-06, | |
| "loss": 0.3778, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 7.787610619469026, | |
| "eval_accuracy": 0.7705864781439808, | |
| "eval_f1": 0.7876889052709178, | |
| "eval_loss": 0.37772777676582336, | |
| "eval_precision": 0.7445993031358885, | |
| "eval_recall": 0.8360719874804382, | |
| "eval_runtime": 53.9925, | |
| "eval_samples_per_second": 186.007, | |
| "eval_steps_per_second": 5.816, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 8.141592920353983, | |
| "grad_norm": 5.115036487579346, | |
| "learning_rate": 1.3176007866273352e-06, | |
| "loss": 0.3832, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 8.141592920353983, | |
| "eval_accuracy": 0.7656078860898138, | |
| "eval_f1": 0.7999320074791773, | |
| "eval_loss": 0.3883359134197235, | |
| "eval_precision": 0.7072437631499849, | |
| "eval_recall": 0.9205790297339593, | |
| "eval_runtime": 53.8543, | |
| "eval_samples_per_second": 186.485, | |
| "eval_steps_per_second": 5.831, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 8.495575221238939, | |
| "grad_norm": 3.359361410140991, | |
| "learning_rate": 1.2782694198623401e-06, | |
| "loss": 0.3723, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 8.495575221238939, | |
| "eval_accuracy": 0.7721796276013143, | |
| "eval_f1": 0.7823853909073616, | |
| "eval_loss": 0.3792983293533325, | |
| "eval_precision": 0.7613846723435764, | |
| "eval_recall": 0.8045774647887324, | |
| "eval_runtime": 53.9123, | |
| "eval_samples_per_second": 186.284, | |
| "eval_steps_per_second": 5.824, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 8.849557522123893, | |
| "grad_norm": 4.47456169128418, | |
| "learning_rate": 1.238938053097345e-06, | |
| "loss": 0.369, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 8.849557522123893, | |
| "eval_accuracy": 0.7759633575624814, | |
| "eval_f1": 0.7839446898405992, | |
| "eval_loss": 0.37068039178848267, | |
| "eval_precision": 0.7698981516408903, | |
| "eval_recall": 0.7985133020344288, | |
| "eval_runtime": 53.9248, | |
| "eval_samples_per_second": 186.241, | |
| "eval_steps_per_second": 5.823, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 9.20353982300885, | |
| "grad_norm": 8.337353706359863, | |
| "learning_rate": 1.19960668633235e-06, | |
| "loss": 0.3683, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 9.20353982300885, | |
| "eval_accuracy": 0.7734740615353978, | |
| "eval_f1": 0.7609540821687507, | |
| "eval_loss": 0.3703781068325043, | |
| "eval_precision": 0.8220204313280364, | |
| "eval_recall": 0.7083333333333334, | |
| "eval_runtime": 53.895, | |
| "eval_samples_per_second": 186.344, | |
| "eval_steps_per_second": 5.826, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 9.557522123893806, | |
| "grad_norm": 8.439238548278809, | |
| "learning_rate": 1.1602753195673548e-06, | |
| "loss": 0.362, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 9.557522123893806, | |
| "eval_accuracy": 0.7749676391516479, | |
| "eval_f1": 0.793644996347699, | |
| "eval_loss": 0.3685624301433563, | |
| "eval_precision": 0.7441780821917808, | |
| "eval_recall": 0.8501564945226917, | |
| "eval_runtime": 53.8072, | |
| "eval_samples_per_second": 186.648, | |
| "eval_steps_per_second": 5.836, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 9.91150442477876, | |
| "grad_norm": 9.439690589904785, | |
| "learning_rate": 1.12094395280236e-06, | |
| "loss": 0.3566, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 9.91150442477876, | |
| "eval_accuracy": 0.7782535099073982, | |
| "eval_f1": 0.7853080111828786, | |
| "eval_loss": 0.3649461269378662, | |
| "eval_precision": 0.7741874168409048, | |
| "eval_recall": 0.7967527386541471, | |
| "eval_runtime": 53.8844, | |
| "eval_samples_per_second": 186.38, | |
| "eval_steps_per_second": 5.827, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 10.265486725663717, | |
| "grad_norm": 9.049650192260742, | |
| "learning_rate": 1.0816125860373648e-06, | |
| "loss": 0.3554, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 10.265486725663717, | |
| "eval_accuracy": 0.7810415214577318, | |
| "eval_f1": 0.7799899949974988, | |
| "eval_loss": 0.3608804941177368, | |
| "eval_precision": 0.7982797460577514, | |
| "eval_recall": 0.7625195618153364, | |
| "eval_runtime": 53.8179, | |
| "eval_samples_per_second": 186.611, | |
| "eval_steps_per_second": 5.834, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 10.619469026548673, | |
| "grad_norm": 5.94489860534668, | |
| "learning_rate": 1.0422812192723695e-06, | |
| "loss": 0.3486, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 10.619469026548673, | |
| "eval_accuracy": 0.7845265358956487, | |
| "eval_f1": 0.7943747624477385, | |
| "eval_loss": 0.3635348081588745, | |
| "eval_precision": 0.7723577235772358, | |
| "eval_recall": 0.8176838810641628, | |
| "eval_runtime": 53.7295, | |
| "eval_samples_per_second": 186.918, | |
| "eval_steps_per_second": 5.844, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 10.973451327433628, | |
| "grad_norm": 22.234378814697266, | |
| "learning_rate": 1.0029498525073746e-06, | |
| "loss": 0.3506, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 10.973451327433628, | |
| "eval_accuracy": 0.7844269640545654, | |
| "eval_f1": 0.7878075075958052, | |
| "eval_loss": 0.3596493601799011, | |
| "eval_precision": 0.7894323315655077, | |
| "eval_recall": 0.786189358372457, | |
| "eval_runtime": 53.8293, | |
| "eval_samples_per_second": 186.571, | |
| "eval_steps_per_second": 5.833, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 11.327433628318584, | |
| "grad_norm": 12.47771167755127, | |
| "learning_rate": 9.636184857423795e-07, | |
| "loss": 0.3441, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 11.327433628318584, | |
| "eval_accuracy": 0.7883102658568157, | |
| "eval_f1": 0.785901309164149, | |
| "eval_loss": 0.35861310362815857, | |
| "eval_precision": 0.8098796180987962, | |
| "eval_recall": 0.763302034428795, | |
| "eval_runtime": 53.7987, | |
| "eval_samples_per_second": 186.677, | |
| "eval_steps_per_second": 5.837, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 11.68141592920354, | |
| "grad_norm": 13.281210899353027, | |
| "learning_rate": 9.242871189773844e-07, | |
| "loss": 0.3416, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 11.68141592920354, | |
| "eval_accuracy": 0.7906004182017325, | |
| "eval_f1": 0.7919263876521223, | |
| "eval_loss": 0.3558104336261749, | |
| "eval_precision": 0.8012012012012012, | |
| "eval_recall": 0.7828638497652582, | |
| "eval_runtime": 53.8593, | |
| "eval_samples_per_second": 186.467, | |
| "eval_steps_per_second": 5.83, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 12.035398230088495, | |
| "grad_norm": 13.627979278564453, | |
| "learning_rate": 8.849557522123894e-07, | |
| "loss": 0.3419, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 12.035398230088495, | |
| "eval_accuracy": 0.7892064124265658, | |
| "eval_f1": 0.8029781293624941, | |
| "eval_loss": 0.35980644822120667, | |
| "eval_precision": 0.7658441327889224, | |
| "eval_recall": 0.8438967136150235, | |
| "eval_runtime": 53.8953, | |
| "eval_samples_per_second": 186.343, | |
| "eval_steps_per_second": 5.826, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 12.389380530973451, | |
| "grad_norm": 23.55402374267578, | |
| "learning_rate": 8.456243854473943e-07, | |
| "loss": 0.3306, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 12.389380530973451, | |
| "eval_accuracy": 0.7884098376978991, | |
| "eval_f1": 0.7980997624703088, | |
| "eval_loss": 0.353899210691452, | |
| "eval_precision": 0.775909846665435, | |
| "eval_recall": 0.8215962441314554, | |
| "eval_runtime": 53.8141, | |
| "eval_samples_per_second": 186.624, | |
| "eval_steps_per_second": 5.835, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 12.743362831858407, | |
| "grad_norm": 18.79817771911621, | |
| "learning_rate": 8.062930186823992e-07, | |
| "loss": 0.335, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 12.743362831858407, | |
| "eval_accuracy": 0.7951807228915663, | |
| "eval_f1": 0.7915695612524065, | |
| "eval_loss": 0.35350722074508667, | |
| "eval_precision": 0.8211057389110784, | |
| "eval_recall": 0.7640845070422535, | |
| "eval_runtime": 53.8902, | |
| "eval_samples_per_second": 186.36, | |
| "eval_steps_per_second": 5.827, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 13.097345132743364, | |
| "grad_norm": 5.215780735015869, | |
| "learning_rate": 7.669616519174042e-07, | |
| "loss": 0.3286, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 13.097345132743364, | |
| "eval_accuracy": 0.7942845763218161, | |
| "eval_f1": 0.8051678611844587, | |
| "eval_loss": 0.35424739122390747, | |
| "eval_precision": 0.7773124544792426, | |
| "eval_recall": 0.835093896713615, | |
| "eval_runtime": 53.8955, | |
| "eval_samples_per_second": 186.342, | |
| "eval_steps_per_second": 5.826, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 13.451327433628318, | |
| "grad_norm": 15.066843032836914, | |
| "learning_rate": 7.27630285152409e-07, | |
| "loss": 0.3239, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 13.451327433628318, | |
| "eval_accuracy": 0.7946828636861496, | |
| "eval_f1": 0.8016544824932667, | |
| "eval_loss": 0.3558191955089569, | |
| "eval_precision": 0.7886071158213475, | |
| "eval_recall": 0.8151408450704225, | |
| "eval_runtime": 53.8826, | |
| "eval_samples_per_second": 186.387, | |
| "eval_steps_per_second": 5.827, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 13.805309734513274, | |
| "grad_norm": 17.407468795776367, | |
| "learning_rate": 6.88298918387414e-07, | |
| "loss": 0.3231, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 13.805309734513274, | |
| "eval_accuracy": 0.7991635965348999, | |
| "eval_f1": 0.7988431235663708, | |
| "eval_loss": 0.351671427488327, | |
| "eval_precision": 0.814852492370295, | |
| "eval_recall": 0.7834507042253521, | |
| "eval_runtime": 53.9126, | |
| "eval_samples_per_second": 186.283, | |
| "eval_steps_per_second": 5.824, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 14.15929203539823, | |
| "grad_norm": 9.669415473937988, | |
| "learning_rate": 6.489675516224188e-07, | |
| "loss": 0.323, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 14.15929203539823, | |
| "eval_accuracy": 0.7965747286667331, | |
| "eval_f1": 0.8164585392148055, | |
| "eval_loss": 0.34596434235572815, | |
| "eval_precision": 0.7549426815085563, | |
| "eval_recall": 0.8888888888888888, | |
| "eval_runtime": 53.8184, | |
| "eval_samples_per_second": 186.609, | |
| "eval_steps_per_second": 5.834, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 14.513274336283185, | |
| "grad_norm": 12.53975772857666, | |
| "learning_rate": 6.096361848574237e-07, | |
| "loss": 0.3158, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 14.513274336283185, | |
| "eval_accuracy": 0.7998605994224833, | |
| "eval_f1": 0.8066192033865692, | |
| "eval_loss": 0.35504859685897827, | |
| "eval_precision": 0.7936387731919727, | |
| "eval_recall": 0.8200312989045383, | |
| "eval_runtime": 53.7599, | |
| "eval_samples_per_second": 186.812, | |
| "eval_steps_per_second": 5.841, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 14.867256637168142, | |
| "grad_norm": 20.021963119506836, | |
| "learning_rate": 5.703048180924287e-07, | |
| "loss": 0.3126, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 14.867256637168142, | |
| "eval_accuracy": 0.8029473264960669, | |
| "eval_f1": 0.7949435291679618, | |
| "eval_loss": 0.34774094820022583, | |
| "eval_precision": 0.8451200705001102, | |
| "eval_recall": 0.7503912363067292, | |
| "eval_runtime": 53.9679, | |
| "eval_samples_per_second": 186.092, | |
| "eval_steps_per_second": 5.818, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 15.221238938053098, | |
| "grad_norm": 27.91417694091797, | |
| "learning_rate": 5.309734513274336e-07, | |
| "loss": 0.3135, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 15.221238938053098, | |
| "eval_accuracy": 0.8033456138604003, | |
| "eval_f1": 0.8052076141631325, | |
| "eval_loss": 0.346435546875, | |
| "eval_precision": 0.8120151183608514, | |
| "eval_recall": 0.7985133020344288, | |
| "eval_runtime": 53.8793, | |
| "eval_samples_per_second": 186.398, | |
| "eval_steps_per_second": 5.828, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 15.575221238938052, | |
| "grad_norm": 9.588032722473145, | |
| "learning_rate": 4.916420845624385e-07, | |
| "loss": 0.3095, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 15.575221238938052, | |
| "eval_accuracy": 0.8035447575425669, | |
| "eval_f1": 0.8067391517288667, | |
| "eval_loss": 0.34088295698165894, | |
| "eval_precision": 0.807926231116343, | |
| "eval_recall": 0.8055555555555556, | |
| "eval_runtime": 53.8032, | |
| "eval_samples_per_second": 186.662, | |
| "eval_steps_per_second": 5.836, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 15.929203539823009, | |
| "grad_norm": 10.44812297821045, | |
| "learning_rate": 4.5231071779744347e-07, | |
| "loss": 0.3099, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 15.929203539823009, | |
| "eval_accuracy": 0.80065717415115, | |
| "eval_f1": 0.8102729340409401, | |
| "eval_loss": 0.3472129702568054, | |
| "eval_precision": 0.7858455882352942, | |
| "eval_recall": 0.8362676056338029, | |
| "eval_runtime": 53.8919, | |
| "eval_samples_per_second": 186.355, | |
| "eval_steps_per_second": 5.826, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 16.283185840707965, | |
| "grad_norm": 10.668062210083008, | |
| "learning_rate": 4.1297935103244836e-07, | |
| "loss": 0.3056, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 16.283185840707965, | |
| "eval_accuracy": 0.8055361943642338, | |
| "eval_f1": 0.7982229569170369, | |
| "eval_loss": 0.3375505208969116, | |
| "eval_precision": 0.8458506678344646, | |
| "eval_recall": 0.7556729264475743, | |
| "eval_runtime": 54.0333, | |
| "eval_samples_per_second": 185.867, | |
| "eval_steps_per_second": 5.811, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 16.63716814159292, | |
| "grad_norm": 62.42535400390625, | |
| "learning_rate": 3.7364798426745325e-07, | |
| "loss": 0.2983, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 16.63716814159292, | |
| "eval_accuracy": 0.8077267748680673, | |
| "eval_f1": 0.8059491508391117, | |
| "eval_loss": 0.34593281149864197, | |
| "eval_precision": 0.8286836123165944, | |
| "eval_recall": 0.7844287949921753, | |
| "eval_runtime": 53.7142, | |
| "eval_samples_per_second": 186.971, | |
| "eval_steps_per_second": 5.846, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 16.991150442477878, | |
| "grad_norm": 20.520849227905273, | |
| "learning_rate": 3.343166175024582e-07, | |
| "loss": 0.2974, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 16.991150442477878, | |
| "eval_accuracy": 0.8077267748680673, | |
| "eval_f1": 0.810295706847431, | |
| "eval_loss": 0.34830448031425476, | |
| "eval_precision": 0.8138938227748175, | |
| "eval_recall": 0.8067292644757433, | |
| "eval_runtime": 53.7486, | |
| "eval_samples_per_second": 186.851, | |
| "eval_steps_per_second": 5.842, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 17.345132743362832, | |
| "grad_norm": 10.991024017333984, | |
| "learning_rate": 2.949852507374631e-07, | |
| "loss": 0.2966, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 17.345132743362832, | |
| "eval_accuracy": 0.8075276311859007, | |
| "eval_f1": 0.8143666570632863, | |
| "eval_loss": 0.35505732893943787, | |
| "eval_precision": 0.7998490850782871, | |
| "eval_recall": 0.8294209702660407, | |
| "eval_runtime": 53.9387, | |
| "eval_samples_per_second": 186.193, | |
| "eval_steps_per_second": 5.821, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 17.699115044247787, | |
| "grad_norm": 16.205684661865234, | |
| "learning_rate": 2.55653883972468e-07, | |
| "loss": 0.2974, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 17.699115044247787, | |
| "eval_accuracy": 0.8090212088021508, | |
| "eval_f1": 0.8085446196845678, | |
| "eval_loss": 0.3468624949455261, | |
| "eval_precision": 0.8255197717081125, | |
| "eval_recall": 0.7922535211267606, | |
| "eval_runtime": 53.9814, | |
| "eval_samples_per_second": 186.045, | |
| "eval_steps_per_second": 5.817, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 18.053097345132745, | |
| "grad_norm": 11.926201820373535, | |
| "learning_rate": 2.1632251720747293e-07, | |
| "loss": 0.2937, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 18.053097345132745, | |
| "eval_accuracy": 0.807925918550234, | |
| "eval_f1": 0.8064224786753638, | |
| "eval_loss": 0.3447153568267822, | |
| "eval_precision": 0.8279414794972182, | |
| "eval_recall": 0.7859937402190923, | |
| "eval_runtime": 53.9266, | |
| "eval_samples_per_second": 186.235, | |
| "eval_steps_per_second": 5.823, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 18.4070796460177, | |
| "grad_norm": 8.203916549682617, | |
| "learning_rate": 1.7699115044247788e-07, | |
| "loss": 0.2902, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 18.4070796460177, | |
| "eval_accuracy": 0.808224634073484, | |
| "eval_f1": 0.8113983548766157, | |
| "eval_loss": 0.34868165850639343, | |
| "eval_precision": 0.8123529411764706, | |
| "eval_recall": 0.8104460093896714, | |
| "eval_runtime": 53.9322, | |
| "eval_samples_per_second": 186.215, | |
| "eval_steps_per_second": 5.822, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 18.761061946902654, | |
| "grad_norm": 15.017583847045898, | |
| "learning_rate": 1.376597836774828e-07, | |
| "loss": 0.2969, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 18.761061946902654, | |
| "eval_accuracy": 0.8092203524843175, | |
| "eval_f1": 0.8131825273010921, | |
| "eval_loss": 0.343014657497406, | |
| "eval_precision": 0.8106531881804043, | |
| "eval_recall": 0.8157276995305164, | |
| "eval_runtime": 53.6911, | |
| "eval_samples_per_second": 187.051, | |
| "eval_steps_per_second": 5.848, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 19.115044247787612, | |
| "grad_norm": 17.386789321899414, | |
| "learning_rate": 9.83284169124877e-08, | |
| "loss": 0.2913, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 19.115044247787612, | |
| "eval_accuracy": 0.8131036542865677, | |
| "eval_f1": 0.8098855464397853, | |
| "eval_loss": 0.33793944120407104, | |
| "eval_precision": 0.8397395505145978, | |
| "eval_recall": 0.7820813771517997, | |
| "eval_runtime": 53.8514, | |
| "eval_samples_per_second": 186.495, | |
| "eval_steps_per_second": 5.831, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 19.469026548672566, | |
| "grad_norm": 10.149681091308594, | |
| "learning_rate": 5.899705014749262e-08, | |
| "loss": 0.2928, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 19.469026548672566, | |
| "eval_accuracy": 0.8094194961664841, | |
| "eval_f1": 0.8097415506958251, | |
| "eval_loss": 0.3402244746685028, | |
| "eval_precision": 0.8231608730800324, | |
| "eval_recall": 0.7967527386541471, | |
| "eval_runtime": 53.8454, | |
| "eval_samples_per_second": 186.516, | |
| "eval_steps_per_second": 5.832, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 19.82300884955752, | |
| "grad_norm": 12.412276268005371, | |
| "learning_rate": 1.966568338249754e-08, | |
| "loss": 0.2894, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 19.82300884955752, | |
| "eval_accuracy": 0.8115105048292343, | |
| "eval_f1": 0.8119972191876055, | |
| "eval_loss": 0.3394980728626251, | |
| "eval_precision": 0.8246923542465201, | |
| "eval_recall": 0.7996870109546166, | |
| "eval_runtime": 53.757, | |
| "eval_samples_per_second": 186.822, | |
| "eval_steps_per_second": 5.841, | |
| "step": 56000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 56500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.8793058646730336e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |