| { | |
| "best_metric": 0.7984189723320159, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-wwv686dm/checkpoint-800", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 800, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 3.582700252532959, | |
| "learning_rate": 2.7292472681730324e-05, | |
| "loss": 0.6088, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 4.199773788452148, | |
| "learning_rate": 2.69469983439869e-05, | |
| "loss": 0.4749, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 5.937159538269043, | |
| "learning_rate": 2.660152400624348e-05, | |
| "loss": 0.5166, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 17.631742477416992, | |
| "learning_rate": 2.6256049668500057e-05, | |
| "loss": 0.6679, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 7.767158508300781, | |
| "learning_rate": 2.5910575330756636e-05, | |
| "loss": 0.4645, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 7.0895304679870605, | |
| "learning_rate": 2.5565100993013214e-05, | |
| "loss": 0.5341, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 23.406824111938477, | |
| "learning_rate": 2.5219626655269793e-05, | |
| "loss": 0.5612, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.8566699624061584, | |
| "learning_rate": 2.4874152317526372e-05, | |
| "loss": 0.524, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 3.7214274406433105, | |
| "learning_rate": 2.4528677979782947e-05, | |
| "loss": 0.4669, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 2.441441297531128, | |
| "learning_rate": 2.4183203642039526e-05, | |
| "loss": 0.4627, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.275, | |
| "grad_norm": 1.745011806488037, | |
| "learning_rate": 2.3837729304296105e-05, | |
| "loss": 0.2824, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 7.687053203582764, | |
| "learning_rate": 2.3492254966552684e-05, | |
| "loss": 0.4334, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.325, | |
| "grad_norm": 6.493716716766357, | |
| "learning_rate": 2.3146780628809263e-05, | |
| "loss": 0.5719, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.3800058960914612, | |
| "learning_rate": 2.2801306291065838e-05, | |
| "loss": 0.3577, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 5.935199737548828, | |
| "learning_rate": 2.2455831953322417e-05, | |
| "loss": 0.5615, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 13.528944969177246, | |
| "learning_rate": 2.2110357615579e-05, | |
| "loss": 0.7921, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.425, | |
| "grad_norm": 1.8475226163864136, | |
| "learning_rate": 2.1764883277835574e-05, | |
| "loss": 0.425, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 6.545809745788574, | |
| "learning_rate": 2.1419408940092153e-05, | |
| "loss": 0.3982, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.475, | |
| "grad_norm": 25.855253219604492, | |
| "learning_rate": 2.107393460234873e-05, | |
| "loss": 0.5213, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 29.630401611328125, | |
| "learning_rate": 2.0728460264605307e-05, | |
| "loss": 0.5541, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.525, | |
| "grad_norm": 24.27338981628418, | |
| "learning_rate": 2.038298592686189e-05, | |
| "loss": 0.3926, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 4.607421398162842, | |
| "learning_rate": 2.0037511589118465e-05, | |
| "loss": 0.3315, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.575, | |
| "grad_norm": 30.252595901489258, | |
| "learning_rate": 1.9692037251375044e-05, | |
| "loss": 0.479, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 30.733047485351562, | |
| "learning_rate": 1.934656291363162e-05, | |
| "loss": 0.7143, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 1.7418993711471558, | |
| "learning_rate": 1.90010885758882e-05, | |
| "loss": 0.2089, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 51.1358528137207, | |
| "learning_rate": 1.865561423814478e-05, | |
| "loss": 0.4703, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.675, | |
| "grad_norm": 0.20157819986343384, | |
| "learning_rate": 1.8310139900401355e-05, | |
| "loss": 0.4402, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.2874705195426941, | |
| "learning_rate": 1.7964665562657934e-05, | |
| "loss": 0.5101, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.725, | |
| "grad_norm": 11.20487117767334, | |
| "learning_rate": 1.7619191224914513e-05, | |
| "loss": 0.4366, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 6.39927864074707, | |
| "learning_rate": 1.727371688717109e-05, | |
| "loss": 0.4827, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.775, | |
| "grad_norm": 0.44607990980148315, | |
| "learning_rate": 1.692824254942767e-05, | |
| "loss": 0.5446, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.5263460874557495, | |
| "learning_rate": 1.6582768211684246e-05, | |
| "loss": 0.3272, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.825, | |
| "grad_norm": 47.18272018432617, | |
| "learning_rate": 1.6237293873940825e-05, | |
| "loss": 0.3534, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 59.770999908447266, | |
| "learning_rate": 1.5891819536197403e-05, | |
| "loss": 0.5308, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 30.36041259765625, | |
| "learning_rate": 1.5546345198453982e-05, | |
| "loss": 0.4749, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 59.363887786865234, | |
| "learning_rate": 1.5200870860710561e-05, | |
| "loss": 0.5161, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.925, | |
| "grad_norm": 0.28430575132369995, | |
| "learning_rate": 1.4855396522967138e-05, | |
| "loss": 0.1804, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.17475275695323944, | |
| "learning_rate": 1.4509922185223717e-05, | |
| "loss": 0.4549, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.975, | |
| "grad_norm": 0.23377352952957153, | |
| "learning_rate": 1.4164447847480294e-05, | |
| "loss": 0.3728, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 26.168912887573242, | |
| "learning_rate": 1.3818973509736873e-05, | |
| "loss": 0.4681, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.845, | |
| "eval_f1": 0.7785714285714286, | |
| "eval_loss": 0.5115727186203003, | |
| "eval_precision": 0.6942675159235668, | |
| "eval_recall": 0.8861788617886179, | |
| "eval_runtime": 1.5294, | |
| "eval_samples_per_second": 261.538, | |
| "eval_steps_per_second": 16.346, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.025, | |
| "grad_norm": 0.34025460481643677, | |
| "learning_rate": 1.347349917199345e-05, | |
| "loss": 0.3656, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 0.2740713357925415, | |
| "learning_rate": 1.3128024834250028e-05, | |
| "loss": 0.2538, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.075, | |
| "grad_norm": 0.42124518752098083, | |
| "learning_rate": 1.2782550496506607e-05, | |
| "loss": 0.3407, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.13944321870803833, | |
| "learning_rate": 1.2437076158763186e-05, | |
| "loss": 0.2763, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 0.40201976895332336, | |
| "learning_rate": 1.2091601821019763e-05, | |
| "loss": 0.291, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 53.71974182128906, | |
| "learning_rate": 1.1746127483276342e-05, | |
| "loss": 0.1759, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.175, | |
| "grad_norm": 0.12816284596920013, | |
| "learning_rate": 1.1400653145532919e-05, | |
| "loss": 0.285, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 23.5927677154541, | |
| "learning_rate": 1.10551788077895e-05, | |
| "loss": 0.8628, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.225, | |
| "grad_norm": 50.016136169433594, | |
| "learning_rate": 1.0709704470046077e-05, | |
| "loss": 0.1292, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 5.992556095123291, | |
| "learning_rate": 1.0364230132302654e-05, | |
| "loss": 0.3215, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.275, | |
| "grad_norm": 5.830613613128662, | |
| "learning_rate": 1.0018755794559232e-05, | |
| "loss": 0.3804, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 110.40708923339844, | |
| "learning_rate": 9.67328145681581e-06, | |
| "loss": 0.189, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.325, | |
| "grad_norm": 0.33183109760284424, | |
| "learning_rate": 9.32780711907239e-06, | |
| "loss": 0.411, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.1316114366054535, | |
| "learning_rate": 8.982332781328967e-06, | |
| "loss": 0.3033, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 20.63913345336914, | |
| "learning_rate": 8.636858443585546e-06, | |
| "loss": 0.3235, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 10.207995414733887, | |
| "learning_rate": 8.291384105842123e-06, | |
| "loss": 0.3278, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.425, | |
| "grad_norm": 11.863898277282715, | |
| "learning_rate": 7.945909768098702e-06, | |
| "loss": 0.6049, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.10968229174613953, | |
| "learning_rate": 7.6004354303552804e-06, | |
| "loss": 0.3248, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.475, | |
| "grad_norm": 9.203657150268555, | |
| "learning_rate": 7.254961092611858e-06, | |
| "loss": 0.4495, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.31232744455337524, | |
| "learning_rate": 6.909486754868436e-06, | |
| "loss": 0.3311, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.525, | |
| "grad_norm": 7.433166980743408, | |
| "learning_rate": 6.564012417125014e-06, | |
| "loss": 0.4362, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.20460058748722076, | |
| "learning_rate": 6.218538079381593e-06, | |
| "loss": 0.4586, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.575, | |
| "grad_norm": 88.86974334716797, | |
| "learning_rate": 5.873063741638171e-06, | |
| "loss": 0.2226, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.2704033851623535, | |
| "learning_rate": 5.52758940389475e-06, | |
| "loss": 0.2187, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 0.2243555188179016, | |
| "learning_rate": 5.182115066151327e-06, | |
| "loss": 0.2537, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 19.40781021118164, | |
| "learning_rate": 4.836640728407905e-06, | |
| "loss": 0.1624, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.675, | |
| "grad_norm": 6.553920745849609, | |
| "learning_rate": 4.4911663906644835e-06, | |
| "loss": 0.3406, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 3.066204071044922, | |
| "learning_rate": 4.1456920529210614e-06, | |
| "loss": 0.6041, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.725, | |
| "grad_norm": 65.46623992919922, | |
| "learning_rate": 3.8002177151776402e-06, | |
| "loss": 0.8913, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.1020149365067482, | |
| "learning_rate": 3.454743377434218e-06, | |
| "loss": 0.3076, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.775, | |
| "grad_norm": 0.8768492937088013, | |
| "learning_rate": 3.1092690396907965e-06, | |
| "loss": 0.174, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 11.6917142868042, | |
| "learning_rate": 2.763794701947375e-06, | |
| "loss": 0.3725, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.825, | |
| "grad_norm": 0.09744672477245331, | |
| "learning_rate": 2.4183203642039524e-06, | |
| "loss": 0.3456, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 0.6613803505897522, | |
| "learning_rate": 2.0728460264605307e-06, | |
| "loss": 0.3638, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 0.1900307685136795, | |
| "learning_rate": 1.727371688717109e-06, | |
| "loss": 0.1932, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 50.69825744628906, | |
| "learning_rate": 1.3818973509736874e-06, | |
| "loss": 0.1641, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.925, | |
| "grad_norm": 0.16248857975006104, | |
| "learning_rate": 1.0364230132302654e-06, | |
| "loss": 0.0771, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 6.616659164428711, | |
| "learning_rate": 6.909486754868437e-07, | |
| "loss": 0.6708, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.975, | |
| "grad_norm": 0.3332100510597229, | |
| "learning_rate": 3.4547433774342186e-07, | |
| "loss": 0.5563, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.16309794783592224, | |
| "learning_rate": 0.0, | |
| "loss": 0.3286, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8725, | |
| "eval_f1": 0.7984189723320159, | |
| "eval_loss": 0.4771757423877716, | |
| "eval_precision": 0.7769230769230769, | |
| "eval_recall": 0.8211382113821138, | |
| "eval_runtime": 1.5707, | |
| "eval_samples_per_second": 254.657, | |
| "eval_steps_per_second": 15.916, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 423630740901888.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 2.7637947019473745e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 2, | |
| "per_device_train_batch_size": 4, | |
| "seed": 8 | |
| } | |
| } | |