| { | |
| "best_metric": 1.89783895, | |
| "best_model_checkpoint": "D:\\_____NEW_NN\\LLM\\MiniCPM-V\\finetune\\output\\phi3-vision-128k-instruct\\v0-20240531-071942\\checkpoint-1476", | |
| "epoch": 4.0, | |
| "eval_steps": 50, | |
| "global_step": 1476, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "acc": 0.47843874, | |
| "epoch": 0.0027100271002710027, | |
| "grad_norm": 0.921875, | |
| "learning_rate": 2.027027027027027e-06, | |
| "loss": 2.52092218, | |
| "memory(GiB)": 12.33, | |
| "step": 1, | |
| "train_speed(iter/s)": 0.066334 | |
| }, | |
| { | |
| "acc": 0.51126236, | |
| "epoch": 0.013550135501355014, | |
| "grad_norm": 0.796875, | |
| "learning_rate": 1.0135135135135135e-05, | |
| "loss": 2.42214346, | |
| "memory(GiB)": 13.38, | |
| "step": 5, | |
| "train_speed(iter/s)": 0.127981 | |
| }, | |
| { | |
| "acc": 0.48636103, | |
| "epoch": 0.02710027100271003, | |
| "grad_norm": 1.09375, | |
| "learning_rate": 2.027027027027027e-05, | |
| "loss": 2.53938828, | |
| "memory(GiB)": 13.38, | |
| "step": 10, | |
| "train_speed(iter/s)": 0.144622 | |
| }, | |
| { | |
| "acc": 0.47292571, | |
| "epoch": 0.04065040650406504, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 3.0405405405405404e-05, | |
| "loss": 2.50403137, | |
| "memory(GiB)": 13.38, | |
| "step": 15, | |
| "train_speed(iter/s)": 0.150329 | |
| }, | |
| { | |
| "acc": 0.52069569, | |
| "epoch": 0.05420054200542006, | |
| "grad_norm": 0.83984375, | |
| "learning_rate": 4.054054054054054e-05, | |
| "loss": 2.28312988, | |
| "memory(GiB)": 13.38, | |
| "step": 20, | |
| "train_speed(iter/s)": 0.152832 | |
| }, | |
| { | |
| "acc": 0.53072515, | |
| "epoch": 0.06775067750677506, | |
| "grad_norm": 0.59765625, | |
| "learning_rate": 5.067567567567567e-05, | |
| "loss": 2.20206394, | |
| "memory(GiB)": 13.38, | |
| "step": 25, | |
| "train_speed(iter/s)": 0.154186 | |
| }, | |
| { | |
| "acc": 0.52423077, | |
| "epoch": 0.08130081300813008, | |
| "grad_norm": 0.89453125, | |
| "learning_rate": 6.081081081081081e-05, | |
| "loss": 2.31712227, | |
| "memory(GiB)": 14.44, | |
| "step": 30, | |
| "train_speed(iter/s)": 0.155403 | |
| }, | |
| { | |
| "acc": 0.54341574, | |
| "epoch": 0.0948509485094851, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 7.094594594594594e-05, | |
| "loss": 2.03825722, | |
| "memory(GiB)": 14.44, | |
| "step": 35, | |
| "train_speed(iter/s)": 0.156383 | |
| }, | |
| { | |
| "acc": 0.56128917, | |
| "epoch": 0.10840108401084012, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 8.108108108108108e-05, | |
| "loss": 1.98176575, | |
| "memory(GiB)": 14.44, | |
| "step": 40, | |
| "train_speed(iter/s)": 0.156973 | |
| }, | |
| { | |
| "acc": 0.53511982, | |
| "epoch": 0.12195121951219512, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 9.121621621621621e-05, | |
| "loss": 2.07999401, | |
| "memory(GiB)": 14.44, | |
| "step": 45, | |
| "train_speed(iter/s)": 0.157395 | |
| }, | |
| { | |
| "acc": 0.56112757, | |
| "epoch": 0.13550135501355012, | |
| "grad_norm": 0.83984375, | |
| "learning_rate": 0.00010135135135135135, | |
| "loss": 1.92188244, | |
| "memory(GiB)": 14.44, | |
| "step": 50, | |
| "train_speed(iter/s)": 0.157924 | |
| }, | |
| { | |
| "epoch": 0.13550135501355012, | |
| "eval_acc": 0.5618983279851376, | |
| "eval_loss": 1.901513695716858, | |
| "eval_runtime": 44.6741, | |
| "eval_samples_per_second": 0.851, | |
| "eval_steps_per_second": 0.851, | |
| "step": 50 | |
| }, | |
| { | |
| "acc": 0.57467785, | |
| "epoch": 0.14905149051490515, | |
| "grad_norm": 0.9921875, | |
| "learning_rate": 0.00011148648648648647, | |
| "loss": 1.92710114, | |
| "memory(GiB)": 15.21, | |
| "step": 55, | |
| "train_speed(iter/s)": 0.140365 | |
| }, | |
| { | |
| "acc": 0.55041471, | |
| "epoch": 0.16260162601626016, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 0.00012162162162162162, | |
| "loss": 2.09731407, | |
| "memory(GiB)": 15.21, | |
| "step": 60, | |
| "train_speed(iter/s)": 0.141894 | |
| }, | |
| { | |
| "acc": 0.53853202, | |
| "epoch": 0.17615176151761516, | |
| "grad_norm": 1.03125, | |
| "learning_rate": 0.00013175675675675675, | |
| "loss": 2.01167774, | |
| "memory(GiB)": 15.21, | |
| "step": 65, | |
| "train_speed(iter/s)": 0.143162 | |
| }, | |
| { | |
| "acc": 0.57625084, | |
| "epoch": 0.1897018970189702, | |
| "grad_norm": 0.81640625, | |
| "learning_rate": 0.00014189189189189188, | |
| "loss": 1.77697067, | |
| "memory(GiB)": 15.21, | |
| "step": 70, | |
| "train_speed(iter/s)": 0.144318 | |
| }, | |
| { | |
| "acc": 0.57468171, | |
| "epoch": 0.2032520325203252, | |
| "grad_norm": 0.96484375, | |
| "learning_rate": 0.00014989300998573466, | |
| "loss": 1.92596684, | |
| "memory(GiB)": 15.21, | |
| "step": 75, | |
| "train_speed(iter/s)": 0.145344 | |
| }, | |
| { | |
| "acc": 0.58551345, | |
| "epoch": 0.21680216802168023, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 0.00014935805991440798, | |
| "loss": 1.85387783, | |
| "memory(GiB)": 15.21, | |
| "step": 80, | |
| "train_speed(iter/s)": 0.146049 | |
| }, | |
| { | |
| "acc": 0.54782548, | |
| "epoch": 0.23035230352303523, | |
| "grad_norm": 1.0078125, | |
| "learning_rate": 0.0001488231098430813, | |
| "loss": 2.04980812, | |
| "memory(GiB)": 15.21, | |
| "step": 85, | |
| "train_speed(iter/s)": 0.146726 | |
| }, | |
| { | |
| "acc": 0.56594706, | |
| "epoch": 0.24390243902439024, | |
| "grad_norm": 1.0, | |
| "learning_rate": 0.0001482881597717546, | |
| "loss": 1.93099308, | |
| "memory(GiB)": 15.21, | |
| "step": 90, | |
| "train_speed(iter/s)": 0.147437 | |
| }, | |
| { | |
| "acc": 0.57289362, | |
| "epoch": 0.25745257452574527, | |
| "grad_norm": 0.8046875, | |
| "learning_rate": 0.00014775320970042795, | |
| "loss": 1.84038963, | |
| "memory(GiB)": 15.58, | |
| "step": 95, | |
| "train_speed(iter/s)": 0.147976 | |
| }, | |
| { | |
| "acc": 0.5938961, | |
| "epoch": 0.27100271002710025, | |
| "grad_norm": 1.015625, | |
| "learning_rate": 0.00014721825962910127, | |
| "loss": 1.71151619, | |
| "memory(GiB)": 15.58, | |
| "step": 100, | |
| "train_speed(iter/s)": 0.148573 | |
| }, | |
| { | |
| "epoch": 0.27100271002710025, | |
| "eval_acc": 0.587907448066205, | |
| "eval_loss": 1.7437644004821777, | |
| "eval_runtime": 44.775, | |
| "eval_samples_per_second": 0.849, | |
| "eval_steps_per_second": 0.849, | |
| "step": 100 | |
| }, | |
| { | |
| "acc": 0.5828393, | |
| "epoch": 0.2845528455284553, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 0.00014668330955777461, | |
| "loss": 1.90272522, | |
| "memory(GiB)": 15.58, | |
| "step": 105, | |
| "train_speed(iter/s)": 0.140241 | |
| }, | |
| { | |
| "acc": 0.57981925, | |
| "epoch": 0.2981029810298103, | |
| "grad_norm": 0.859375, | |
| "learning_rate": 0.00014614835948644793, | |
| "loss": 1.88997154, | |
| "memory(GiB)": 15.58, | |
| "step": 110, | |
| "train_speed(iter/s)": 0.141057 | |
| }, | |
| { | |
| "acc": 0.56191244, | |
| "epoch": 0.3116531165311653, | |
| "grad_norm": 0.94140625, | |
| "learning_rate": 0.00014561340941512125, | |
| "loss": 1.92102909, | |
| "memory(GiB)": 15.58, | |
| "step": 115, | |
| "train_speed(iter/s)": 0.141824 | |
| }, | |
| { | |
| "acc": 0.57420607, | |
| "epoch": 0.3252032520325203, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 0.00014507845934379456, | |
| "loss": 1.80924416, | |
| "memory(GiB)": 15.58, | |
| "step": 120, | |
| "train_speed(iter/s)": 0.14256 | |
| }, | |
| { | |
| "acc": 0.57703466, | |
| "epoch": 0.33875338753387535, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 0.00014454350927246788, | |
| "loss": 1.85804043, | |
| "memory(GiB)": 15.58, | |
| "step": 125, | |
| "train_speed(iter/s)": 0.143173 | |
| }, | |
| { | |
| "acc": 0.58581357, | |
| "epoch": 0.3523035230352303, | |
| "grad_norm": 0.96484375, | |
| "learning_rate": 0.0001440085592011412, | |
| "loss": 1.75103779, | |
| "memory(GiB)": 15.58, | |
| "step": 130, | |
| "train_speed(iter/s)": 0.143759 | |
| }, | |
| { | |
| "acc": 0.59179254, | |
| "epoch": 0.36585365853658536, | |
| "grad_norm": 0.93359375, | |
| "learning_rate": 0.00014347360912981454, | |
| "loss": 1.67790794, | |
| "memory(GiB)": 15.58, | |
| "step": 135, | |
| "train_speed(iter/s)": 0.14428 | |
| }, | |
| { | |
| "acc": 0.59449296, | |
| "epoch": 0.3794037940379404, | |
| "grad_norm": 0.74609375, | |
| "learning_rate": 0.00014293865905848786, | |
| "loss": 1.72561359, | |
| "memory(GiB)": 15.58, | |
| "step": 140, | |
| "train_speed(iter/s)": 0.144799 | |
| }, | |
| { | |
| "acc": 0.57711525, | |
| "epoch": 0.39295392953929537, | |
| "grad_norm": 1.109375, | |
| "learning_rate": 0.0001424037089871612, | |
| "loss": 1.79506092, | |
| "memory(GiB)": 15.58, | |
| "step": 145, | |
| "train_speed(iter/s)": 0.14532 | |
| }, | |
| { | |
| "acc": 0.57541366, | |
| "epoch": 0.4065040650406504, | |
| "grad_norm": 5.46875, | |
| "learning_rate": 0.00014186875891583452, | |
| "loss": 1.80377541, | |
| "memory(GiB)": 15.58, | |
| "step": 150, | |
| "train_speed(iter/s)": 0.145757 | |
| }, | |
| { | |
| "epoch": 0.4065040650406504, | |
| "eval_acc": 0.5983786522546867, | |
| "eval_loss": 1.676965594291687, | |
| "eval_runtime": 44.3227, | |
| "eval_samples_per_second": 0.857, | |
| "eval_steps_per_second": 0.857, | |
| "step": 150 | |
| }, | |
| { | |
| "acc": 0.58702893, | |
| "epoch": 0.42005420054200543, | |
| "grad_norm": 1.0625, | |
| "learning_rate": 0.00014133380884450783, | |
| "loss": 1.86865826, | |
| "memory(GiB)": 15.58, | |
| "step": 155, | |
| "train_speed(iter/s)": 0.140358 | |
| }, | |
| { | |
| "acc": 0.58541198, | |
| "epoch": 0.43360433604336046, | |
| "grad_norm": 1.015625, | |
| "learning_rate": 0.00014079885877318115, | |
| "loss": 1.82644196, | |
| "memory(GiB)": 15.58, | |
| "step": 160, | |
| "train_speed(iter/s)": 0.140936 | |
| }, | |
| { | |
| "acc": 0.58856125, | |
| "epoch": 0.44715447154471544, | |
| "grad_norm": 0.9921875, | |
| "learning_rate": 0.00014026390870185447, | |
| "loss": 1.68714104, | |
| "memory(GiB)": 15.58, | |
| "step": 165, | |
| "train_speed(iter/s)": 0.141495 | |
| }, | |
| { | |
| "acc": 0.59458299, | |
| "epoch": 0.46070460704607047, | |
| "grad_norm": 0.91015625, | |
| "learning_rate": 0.0001397289586305278, | |
| "loss": 1.72063637, | |
| "memory(GiB)": 15.58, | |
| "step": 170, | |
| "train_speed(iter/s)": 0.142005 | |
| }, | |
| { | |
| "acc": 0.57354069, | |
| "epoch": 0.4742547425474255, | |
| "grad_norm": 0.7734375, | |
| "learning_rate": 0.00013919400855920113, | |
| "loss": 1.83291931, | |
| "memory(GiB)": 15.58, | |
| "step": 175, | |
| "train_speed(iter/s)": 0.142509 | |
| }, | |
| { | |
| "acc": 0.58851786, | |
| "epoch": 0.4878048780487805, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 0.00013865905848787447, | |
| "loss": 1.70540199, | |
| "memory(GiB)": 15.58, | |
| "step": 180, | |
| "train_speed(iter/s)": 0.143002 | |
| }, | |
| { | |
| "acc": 0.59666262, | |
| "epoch": 0.5013550135501355, | |
| "grad_norm": 0.87890625, | |
| "learning_rate": 0.0001381241084165478, | |
| "loss": 1.70772285, | |
| "memory(GiB)": 15.58, | |
| "step": 185, | |
| "train_speed(iter/s)": 0.143463 | |
| }, | |
| { | |
| "acc": 0.58001013, | |
| "epoch": 0.5149051490514905, | |
| "grad_norm": 0.890625, | |
| "learning_rate": 0.0001375891583452211, | |
| "loss": 1.73730106, | |
| "memory(GiB)": 15.58, | |
| "step": 190, | |
| "train_speed(iter/s)": 0.143883 | |
| }, | |
| { | |
| "acc": 0.5934463, | |
| "epoch": 0.5284552845528455, | |
| "grad_norm": 1.421875, | |
| "learning_rate": 0.00013705420827389442, | |
| "loss": 1.68941402, | |
| "memory(GiB)": 15.58, | |
| "step": 195, | |
| "train_speed(iter/s)": 0.144297 | |
| }, | |
| { | |
| "acc": 0.58476119, | |
| "epoch": 0.5420054200542005, | |
| "grad_norm": 0.91015625, | |
| "learning_rate": 0.00013651925820256774, | |
| "loss": 1.74896088, | |
| "memory(GiB)": 15.58, | |
| "step": 200, | |
| "train_speed(iter/s)": 0.144655 | |
| }, | |
| { | |
| "epoch": 0.5420054200542005, | |
| "eval_acc": 0.6105387603445364, | |
| "eval_loss": 1.6297248601913452, | |
| "eval_runtime": 44.3331, | |
| "eval_samples_per_second": 0.857, | |
| "eval_steps_per_second": 0.857, | |
| "step": 200 | |
| }, | |
| { | |
| "acc": 0.58323898, | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 0.79296875, | |
| "learning_rate": 0.00013598430813124105, | |
| "loss": 1.74196358, | |
| "memory(GiB)": 15.58, | |
| "step": 205, | |
| "train_speed(iter/s)": 0.140612 | |
| }, | |
| { | |
| "acc": 0.59854908, | |
| "epoch": 0.5691056910569106, | |
| "grad_norm": 1.109375, | |
| "learning_rate": 0.0001354493580599144, | |
| "loss": 1.6279623, | |
| "memory(GiB)": 15.58, | |
| "step": 210, | |
| "train_speed(iter/s)": 0.141053 | |
| }, | |
| { | |
| "acc": 0.58306313, | |
| "epoch": 0.5826558265582655, | |
| "grad_norm": 0.98828125, | |
| "learning_rate": 0.00013491440798858771, | |
| "loss": 1.85492191, | |
| "memory(GiB)": 15.58, | |
| "step": 215, | |
| "train_speed(iter/s)": 0.141461 | |
| }, | |
| { | |
| "acc": 0.58454275, | |
| "epoch": 0.5962059620596206, | |
| "grad_norm": 0.890625, | |
| "learning_rate": 0.00013437945791726106, | |
| "loss": 1.75104046, | |
| "memory(GiB)": 15.58, | |
| "step": 220, | |
| "train_speed(iter/s)": 0.141869 | |
| }, | |
| { | |
| "acc": 0.60898943, | |
| "epoch": 0.6097560975609756, | |
| "grad_norm": 0.828125, | |
| "learning_rate": 0.00013384450784593437, | |
| "loss": 1.57786808, | |
| "memory(GiB)": 15.58, | |
| "step": 225, | |
| "train_speed(iter/s)": 0.142237 | |
| }, | |
| { | |
| "acc": 0.58954048, | |
| "epoch": 0.6233062330623306, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 0.0001333095577746077, | |
| "loss": 1.72581081, | |
| "memory(GiB)": 15.58, | |
| "step": 230, | |
| "train_speed(iter/s)": 0.142622 | |
| }, | |
| { | |
| "acc": 0.59608021, | |
| "epoch": 0.6368563685636857, | |
| "grad_norm": 1.2265625, | |
| "learning_rate": 0.000132774607703281, | |
| "loss": 1.70160866, | |
| "memory(GiB)": 15.58, | |
| "step": 235, | |
| "train_speed(iter/s)": 0.142983 | |
| }, | |
| { | |
| "acc": 0.57084417, | |
| "epoch": 0.6504065040650406, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 0.00013223965763195432, | |
| "loss": 1.81941319, | |
| "memory(GiB)": 15.58, | |
| "step": 240, | |
| "train_speed(iter/s)": 0.143329 | |
| }, | |
| { | |
| "acc": 0.61476159, | |
| "epoch": 0.6639566395663956, | |
| "grad_norm": 0.9453125, | |
| "learning_rate": 0.00013170470756062767, | |
| "loss": 1.68505306, | |
| "memory(GiB)": 15.58, | |
| "step": 245, | |
| "train_speed(iter/s)": 0.143652 | |
| }, | |
| { | |
| "acc": 0.5995626, | |
| "epoch": 0.6775067750677507, | |
| "grad_norm": 0.9140625, | |
| "learning_rate": 0.00013116975748930098, | |
| "loss": 1.77789631, | |
| "memory(GiB)": 15.58, | |
| "step": 250, | |
| "train_speed(iter/s)": 0.144002 | |
| }, | |
| { | |
| "epoch": 0.6775067750677507, | |
| "eval_acc": 0.61087654112481, | |
| "eval_loss": 1.6108067035675049, | |
| "eval_runtime": 44.1094, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 250 | |
| }, | |
| { | |
| "acc": 0.62165804, | |
| "epoch": 0.6910569105691057, | |
| "grad_norm": 1.6171875, | |
| "learning_rate": 0.0001306348074179743, | |
| "loss": 1.57510653, | |
| "memory(GiB)": 15.58, | |
| "step": 255, | |
| "train_speed(iter/s)": 0.140805 | |
| }, | |
| { | |
| "acc": 0.59346585, | |
| "epoch": 0.7046070460704607, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 0.00013009985734664764, | |
| "loss": 1.76646061, | |
| "memory(GiB)": 15.58, | |
| "step": 260, | |
| "train_speed(iter/s)": 0.141165 | |
| }, | |
| { | |
| "acc": 0.60632119, | |
| "epoch": 0.7181571815718157, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 0.00012956490727532096, | |
| "loss": 1.59940271, | |
| "memory(GiB)": 15.95, | |
| "step": 265, | |
| "train_speed(iter/s)": 0.141502 | |
| }, | |
| { | |
| "acc": 0.59642992, | |
| "epoch": 0.7317073170731707, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 0.00012902995720399428, | |
| "loss": 1.6549921, | |
| "memory(GiB)": 15.95, | |
| "step": 270, | |
| "train_speed(iter/s)": 0.141855 | |
| }, | |
| { | |
| "acc": 0.60280037, | |
| "epoch": 0.7452574525745257, | |
| "grad_norm": 1.0625, | |
| "learning_rate": 0.0001284950071326676, | |
| "loss": 1.63179569, | |
| "memory(GiB)": 15.95, | |
| "step": 275, | |
| "train_speed(iter/s)": 0.142164 | |
| }, | |
| { | |
| "acc": 0.594034, | |
| "epoch": 0.7588075880758808, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 0.00012796005706134094, | |
| "loss": 1.61152973, | |
| "memory(GiB)": 15.95, | |
| "step": 280, | |
| "train_speed(iter/s)": 0.142474 | |
| }, | |
| { | |
| "acc": 0.59763761, | |
| "epoch": 0.7723577235772358, | |
| "grad_norm": 1.0859375, | |
| "learning_rate": 0.00012742510699001425, | |
| "loss": 1.66443863, | |
| "memory(GiB)": 15.95, | |
| "step": 285, | |
| "train_speed(iter/s)": 0.142791 | |
| }, | |
| { | |
| "acc": 0.58316121, | |
| "epoch": 0.7859078590785907, | |
| "grad_norm": 0.73828125, | |
| "learning_rate": 0.00012689015691868757, | |
| "loss": 1.77028389, | |
| "memory(GiB)": 15.95, | |
| "step": 290, | |
| "train_speed(iter/s)": 0.143067 | |
| }, | |
| { | |
| "acc": 0.60665183, | |
| "epoch": 0.7994579945799458, | |
| "grad_norm": 1.25, | |
| "learning_rate": 0.00012635520684736091, | |
| "loss": 1.64849663, | |
| "memory(GiB)": 15.95, | |
| "step": 295, | |
| "train_speed(iter/s)": 0.143366 | |
| }, | |
| { | |
| "acc": 0.55526123, | |
| "epoch": 0.8130081300813008, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 0.00012582025677603423, | |
| "loss": 1.86367321, | |
| "memory(GiB)": 15.95, | |
| "step": 300, | |
| "train_speed(iter/s)": 0.143649 | |
| }, | |
| { | |
| "epoch": 0.8130081300813008, | |
| "eval_acc": 0.6129032258064516, | |
| "eval_loss": 1.5865528583526611, | |
| "eval_runtime": 44.0975, | |
| "eval_samples_per_second": 0.862, | |
| "eval_steps_per_second": 0.862, | |
| "step": 300 | |
| }, | |
| { | |
| "acc": 0.6019598, | |
| "epoch": 0.8265582655826558, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 0.00012528530670470755, | |
| "loss": 1.71853142, | |
| "memory(GiB)": 15.95, | |
| "step": 305, | |
| "train_speed(iter/s)": 0.14098 | |
| }, | |
| { | |
| "acc": 0.6009192, | |
| "epoch": 0.8401084010840109, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 0.00012475035663338086, | |
| "loss": 1.72751312, | |
| "memory(GiB)": 16.34, | |
| "step": 310, | |
| "train_speed(iter/s)": 0.141284 | |
| }, | |
| { | |
| "acc": 0.58907671, | |
| "epoch": 0.8536585365853658, | |
| "grad_norm": 0.98046875, | |
| "learning_rate": 0.0001242154065620542, | |
| "loss": 1.71860523, | |
| "memory(GiB)": 16.34, | |
| "step": 315, | |
| "train_speed(iter/s)": 0.141562 | |
| }, | |
| { | |
| "acc": 0.62246222, | |
| "epoch": 0.8672086720867209, | |
| "grad_norm": 1.0234375, | |
| "learning_rate": 0.00012368045649072752, | |
| "loss": 1.58847914, | |
| "memory(GiB)": 16.34, | |
| "step": 320, | |
| "train_speed(iter/s)": 0.141848 | |
| }, | |
| { | |
| "acc": 0.61927052, | |
| "epoch": 0.8807588075880759, | |
| "grad_norm": 1.59375, | |
| "learning_rate": 0.00012314550641940084, | |
| "loss": 1.54207745, | |
| "memory(GiB)": 16.34, | |
| "step": 325, | |
| "train_speed(iter/s)": 0.142133 | |
| }, | |
| { | |
| "acc": 0.60672359, | |
| "epoch": 0.8943089430894309, | |
| "grad_norm": 2.140625, | |
| "learning_rate": 0.00012261055634807416, | |
| "loss": 1.71789646, | |
| "memory(GiB)": 16.34, | |
| "step": 330, | |
| "train_speed(iter/s)": 0.142392 | |
| }, | |
| { | |
| "acc": 0.63249903, | |
| "epoch": 0.907859078590786, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 0.0001220756062767475, | |
| "loss": 1.46934195, | |
| "memory(GiB)": 16.34, | |
| "step": 335, | |
| "train_speed(iter/s)": 0.142658 | |
| }, | |
| { | |
| "acc": 0.60973873, | |
| "epoch": 0.9214092140921409, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 0.00012154065620542082, | |
| "loss": 1.67951736, | |
| "memory(GiB)": 16.34, | |
| "step": 340, | |
| "train_speed(iter/s)": 0.142914 | |
| }, | |
| { | |
| "acc": 0.58657136, | |
| "epoch": 0.9349593495934959, | |
| "grad_norm": 1.2734375, | |
| "learning_rate": 0.00012100570613409413, | |
| "loss": 1.68692303, | |
| "memory(GiB)": 16.34, | |
| "step": 345, | |
| "train_speed(iter/s)": 0.143167 | |
| }, | |
| { | |
| "acc": 0.59965162, | |
| "epoch": 0.948509485094851, | |
| "grad_norm": 1.421875, | |
| "learning_rate": 0.00012047075606276746, | |
| "loss": 1.71949959, | |
| "memory(GiB)": 16.34, | |
| "step": 350, | |
| "train_speed(iter/s)": 0.143387 | |
| }, | |
| { | |
| "epoch": 0.948509485094851, | |
| "eval_acc": 0.6171254855598717, | |
| "eval_loss": 1.562721610069275, | |
| "eval_runtime": 44.0566, | |
| "eval_samples_per_second": 0.863, | |
| "eval_steps_per_second": 0.863, | |
| "step": 350 | |
| }, | |
| { | |
| "acc": 0.59322553, | |
| "epoch": 0.962059620596206, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 0.0001199358059914408, | |
| "loss": 1.66712799, | |
| "memory(GiB)": 16.34, | |
| "step": 355, | |
| "train_speed(iter/s)": 0.14111 | |
| }, | |
| { | |
| "acc": 0.59898105, | |
| "epoch": 0.975609756097561, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 0.00011940085592011411, | |
| "loss": 1.66754684, | |
| "memory(GiB)": 16.34, | |
| "step": 360, | |
| "train_speed(iter/s)": 0.141373 | |
| }, | |
| { | |
| "acc": 0.60211391, | |
| "epoch": 0.989159891598916, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 0.00011886590584878744, | |
| "loss": 1.67542057, | |
| "memory(GiB)": 16.34, | |
| "step": 365, | |
| "train_speed(iter/s)": 0.141638 | |
| }, | |
| { | |
| "acc": 0.63076615, | |
| "epoch": 1.002710027100271, | |
| "grad_norm": 0.8515625, | |
| "learning_rate": 0.00011833095577746076, | |
| "loss": 1.51196938, | |
| "memory(GiB)": 16.34, | |
| "step": 370, | |
| "train_speed(iter/s)": 0.141355 | |
| }, | |
| { | |
| "acc": 0.63542643, | |
| "epoch": 1.016260162601626, | |
| "grad_norm": 0.828125, | |
| "learning_rate": 0.00011779600570613407, | |
| "loss": 1.38132334, | |
| "memory(GiB)": 16.34, | |
| "step": 375, | |
| "train_speed(iter/s)": 0.141603 | |
| }, | |
| { | |
| "acc": 0.66358423, | |
| "epoch": 1.029810298102981, | |
| "grad_norm": 0.859375, | |
| "learning_rate": 0.00011726105563480742, | |
| "loss": 1.41512909, | |
| "memory(GiB)": 16.34, | |
| "step": 380, | |
| "train_speed(iter/s)": 0.141845 | |
| }, | |
| { | |
| "acc": 0.63692493, | |
| "epoch": 1.043360433604336, | |
| "grad_norm": 1.0859375, | |
| "learning_rate": 0.00011672610556348073, | |
| "loss": 1.38078823, | |
| "memory(GiB)": 16.34, | |
| "step": 385, | |
| "train_speed(iter/s)": 0.142082 | |
| }, | |
| { | |
| "acc": 0.6641499, | |
| "epoch": 1.056910569105691, | |
| "grad_norm": 1.125, | |
| "learning_rate": 0.00011619115549215406, | |
| "loss": 1.3458046, | |
| "memory(GiB)": 16.34, | |
| "step": 390, | |
| "train_speed(iter/s)": 0.142305 | |
| }, | |
| { | |
| "acc": 0.67276783, | |
| "epoch": 1.070460704607046, | |
| "grad_norm": 0.98828125, | |
| "learning_rate": 0.00011565620542082738, | |
| "loss": 1.2076004, | |
| "memory(GiB)": 16.34, | |
| "step": 395, | |
| "train_speed(iter/s)": 0.142533 | |
| }, | |
| { | |
| "acc": 0.65821433, | |
| "epoch": 1.084010840108401, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 0.0001151212553495007, | |
| "loss": 1.38641891, | |
| "memory(GiB)": 16.34, | |
| "step": 400, | |
| "train_speed(iter/s)": 0.142742 | |
| }, | |
| { | |
| "epoch": 1.084010840108401, | |
| "eval_acc": 0.6162810336091876, | |
| "eval_loss": 1.5887514352798462, | |
| "eval_runtime": 44.1624, | |
| "eval_samples_per_second": 0.86, | |
| "eval_steps_per_second": 0.86, | |
| "step": 400 | |
| }, | |
| { | |
| "acc": 0.63352804, | |
| "epoch": 1.0975609756097562, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 0.00011458630527817403, | |
| "loss": 1.44696302, | |
| "memory(GiB)": 16.34, | |
| "step": 405, | |
| "train_speed(iter/s)": 0.140759 | |
| }, | |
| { | |
| "acc": 0.61955671, | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 0.00011405135520684734, | |
| "loss": 1.48915672, | |
| "memory(GiB)": 16.34, | |
| "step": 410, | |
| "train_speed(iter/s)": 0.140987 | |
| }, | |
| { | |
| "acc": 0.69259667, | |
| "epoch": 1.1246612466124661, | |
| "grad_norm": 0.796875, | |
| "learning_rate": 0.00011351640513552069, | |
| "loss": 1.22069292, | |
| "memory(GiB)": 16.34, | |
| "step": 415, | |
| "train_speed(iter/s)": 0.141191 | |
| }, | |
| { | |
| "acc": 0.67346158, | |
| "epoch": 1.1382113821138211, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 0.000112981455064194, | |
| "loss": 1.36795778, | |
| "memory(GiB)": 16.34, | |
| "step": 420, | |
| "train_speed(iter/s)": 0.141417 | |
| }, | |
| { | |
| "acc": 0.67346702, | |
| "epoch": 1.151761517615176, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 0.00011244650499286732, | |
| "loss": 1.27111759, | |
| "memory(GiB)": 16.34, | |
| "step": 425, | |
| "train_speed(iter/s)": 0.141634 | |
| }, | |
| { | |
| "acc": 0.61625972, | |
| "epoch": 1.165311653116531, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 0.00011191155492154065, | |
| "loss": 1.42474003, | |
| "memory(GiB)": 16.34, | |
| "step": 430, | |
| "train_speed(iter/s)": 0.141853 | |
| }, | |
| { | |
| "acc": 0.65880041, | |
| "epoch": 1.1788617886178863, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 0.00011137660485021397, | |
| "loss": 1.33981524, | |
| "memory(GiB)": 16.34, | |
| "step": 435, | |
| "train_speed(iter/s)": 0.142048 | |
| }, | |
| { | |
| "acc": 0.63503499, | |
| "epoch": 1.1924119241192412, | |
| "grad_norm": 1.2265625, | |
| "learning_rate": 0.00011084165477888728, | |
| "loss": 1.44122829, | |
| "memory(GiB)": 16.34, | |
| "step": 440, | |
| "train_speed(iter/s)": 0.142254 | |
| }, | |
| { | |
| "acc": 0.61518383, | |
| "epoch": 1.2059620596205962, | |
| "grad_norm": 1.4765625, | |
| "learning_rate": 0.00011030670470756061, | |
| "loss": 1.54818697, | |
| "memory(GiB)": 16.34, | |
| "step": 445, | |
| "train_speed(iter/s)": 0.142446 | |
| }, | |
| { | |
| "acc": 0.68010144, | |
| "epoch": 1.2195121951219512, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 0.00010977175463623394, | |
| "loss": 1.28560524, | |
| "memory(GiB)": 16.34, | |
| "step": 450, | |
| "train_speed(iter/s)": 0.142639 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "eval_acc": 0.6230366492146597, | |
| "eval_loss": 1.5483555793762207, | |
| "eval_runtime": 44.0588, | |
| "eval_samples_per_second": 0.862, | |
| "eval_steps_per_second": 0.862, | |
| "step": 450 | |
| }, | |
| { | |
| "acc": 0.64718337, | |
| "epoch": 1.2330623306233062, | |
| "grad_norm": 1.4375, | |
| "learning_rate": 0.00010923680456490727, | |
| "loss": 1.3778326, | |
| "memory(GiB)": 16.34, | |
| "step": 455, | |
| "train_speed(iter/s)": 0.140879 | |
| }, | |
| { | |
| "acc": 0.68637676, | |
| "epoch": 1.2466124661246614, | |
| "grad_norm": 1.453125, | |
| "learning_rate": 0.00010870185449358059, | |
| "loss": 1.21860657, | |
| "memory(GiB)": 16.34, | |
| "step": 460, | |
| "train_speed(iter/s)": 0.141097 | |
| }, | |
| { | |
| "acc": 0.63144946, | |
| "epoch": 1.2601626016260163, | |
| "grad_norm": 1.0703125, | |
| "learning_rate": 0.00010816690442225392, | |
| "loss": 1.38577023, | |
| "memory(GiB)": 16.34, | |
| "step": 465, | |
| "train_speed(iter/s)": 0.141294 | |
| }, | |
| { | |
| "acc": 0.66895781, | |
| "epoch": 1.2737127371273713, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 0.00010763195435092724, | |
| "loss": 1.29167385, | |
| "memory(GiB)": 16.34, | |
| "step": 470, | |
| "train_speed(iter/s)": 0.141488 | |
| }, | |
| { | |
| "acc": 0.6553885, | |
| "epoch": 1.2872628726287263, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 0.00010709700427960055, | |
| "loss": 1.35011473, | |
| "memory(GiB)": 16.34, | |
| "step": 475, | |
| "train_speed(iter/s)": 0.141687 | |
| }, | |
| { | |
| "acc": 0.65576148, | |
| "epoch": 1.3008130081300813, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 0.00010656205420827388, | |
| "loss": 1.39116755, | |
| "memory(GiB)": 16.34, | |
| "step": 480, | |
| "train_speed(iter/s)": 0.141863 | |
| }, | |
| { | |
| "acc": 0.62025633, | |
| "epoch": 1.3143631436314362, | |
| "grad_norm": 1.375, | |
| "learning_rate": 0.00010602710413694721, | |
| "loss": 1.55190315, | |
| "memory(GiB)": 16.34, | |
| "step": 485, | |
| "train_speed(iter/s)": 0.142058 | |
| }, | |
| { | |
| "acc": 0.65667233, | |
| "epoch": 1.3279132791327912, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 0.00010549215406562054, | |
| "loss": 1.30497828, | |
| "memory(GiB)": 16.34, | |
| "step": 490, | |
| "train_speed(iter/s)": 0.142237 | |
| }, | |
| { | |
| "acc": 0.6458045, | |
| "epoch": 1.3414634146341464, | |
| "grad_norm": 2.421875, | |
| "learning_rate": 0.00010495720399429386, | |
| "loss": 1.40175285, | |
| "memory(GiB)": 16.34, | |
| "step": 495, | |
| "train_speed(iter/s)": 0.14241 | |
| }, | |
| { | |
| "acc": 0.65004997, | |
| "epoch": 1.3550135501355014, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 0.00010442225392296718, | |
| "loss": 1.31625471, | |
| "memory(GiB)": 16.34, | |
| "step": 500, | |
| "train_speed(iter/s)": 0.142583 | |
| }, | |
| { | |
| "epoch": 1.3550135501355014, | |
| "eval_acc": 0.61932106063165, | |
| "eval_loss": 1.535814642906189, | |
| "eval_runtime": 44.07, | |
| "eval_samples_per_second": 0.862, | |
| "eval_steps_per_second": 0.862, | |
| "step": 500 | |
| }, | |
| { | |
| "acc": 0.65512385, | |
| "epoch": 1.3685636856368564, | |
| "grad_norm": 0.9375, | |
| "learning_rate": 0.00010388730385164051, | |
| "loss": 1.2989893, | |
| "memory(GiB)": 16.34, | |
| "step": 505, | |
| "train_speed(iter/s)": 0.140995 | |
| }, | |
| { | |
| "acc": 0.65444102, | |
| "epoch": 1.3821138211382114, | |
| "grad_norm": 1.5625, | |
| "learning_rate": 0.00010335235378031382, | |
| "loss": 1.32140932, | |
| "memory(GiB)": 16.34, | |
| "step": 510, | |
| "train_speed(iter/s)": 0.141172 | |
| }, | |
| { | |
| "acc": 0.66242504, | |
| "epoch": 1.3956639566395663, | |
| "grad_norm": 0.9140625, | |
| "learning_rate": 0.00010281740370898714, | |
| "loss": 1.34259834, | |
| "memory(GiB)": 16.34, | |
| "step": 515, | |
| "train_speed(iter/s)": 0.141336 | |
| }, | |
| { | |
| "acc": 0.62716827, | |
| "epoch": 1.4092140921409215, | |
| "grad_norm": 1.4296875, | |
| "learning_rate": 0.00010228245363766048, | |
| "loss": 1.46508284, | |
| "memory(GiB)": 16.34, | |
| "step": 520, | |
| "train_speed(iter/s)": 0.141508 | |
| }, | |
| { | |
| "acc": 0.61702657, | |
| "epoch": 1.4227642276422765, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 0.0001017475035663338, | |
| "loss": 1.43188276, | |
| "memory(GiB)": 16.34, | |
| "step": 525, | |
| "train_speed(iter/s)": 0.141681 | |
| }, | |
| { | |
| "acc": 0.63238263, | |
| "epoch": 1.4363143631436315, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 0.00010121255349500713, | |
| "loss": 1.52214499, | |
| "memory(GiB)": 16.34, | |
| "step": 530, | |
| "train_speed(iter/s)": 0.141854 | |
| }, | |
| { | |
| "acc": 0.6323854, | |
| "epoch": 1.4498644986449865, | |
| "grad_norm": 1.375, | |
| "learning_rate": 0.00010067760342368045, | |
| "loss": 1.38910236, | |
| "memory(GiB)": 16.34, | |
| "step": 535, | |
| "train_speed(iter/s)": 0.142019 | |
| }, | |
| { | |
| "acc": 0.60772176, | |
| "epoch": 1.4634146341463414, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 0.00010014265335235376, | |
| "loss": 1.46768923, | |
| "memory(GiB)": 16.34, | |
| "step": 540, | |
| "train_speed(iter/s)": 0.14218 | |
| }, | |
| { | |
| "acc": 0.67315965, | |
| "epoch": 1.4769647696476964, | |
| "grad_norm": 1.34375, | |
| "learning_rate": 9.96077032810271e-05, | |
| "loss": 1.25136938, | |
| "memory(GiB)": 16.34, | |
| "step": 545, | |
| "train_speed(iter/s)": 0.142338 | |
| }, | |
| { | |
| "acc": 0.66408758, | |
| "epoch": 1.4905149051490514, | |
| "grad_norm": 2.046875, | |
| "learning_rate": 9.907275320970041e-05, | |
| "loss": 1.25655928, | |
| "memory(GiB)": 16.34, | |
| "step": 550, | |
| "train_speed(iter/s)": 0.142495 | |
| }, | |
| { | |
| "epoch": 1.4905149051490514, | |
| "eval_acc": 0.6254011146765749, | |
| "eval_loss": 1.5346177816390991, | |
| "eval_runtime": 44.1805, | |
| "eval_samples_per_second": 0.86, | |
| "eval_steps_per_second": 0.86, | |
| "step": 550 | |
| }, | |
| { | |
| "acc": 0.68655195, | |
| "epoch": 1.5040650406504064, | |
| "grad_norm": 1.125, | |
| "learning_rate": 9.853780313837375e-05, | |
| "loss": 1.08906736, | |
| "memory(GiB)": 16.34, | |
| "step": 555, | |
| "train_speed(iter/s)": 0.141037 | |
| }, | |
| { | |
| "acc": 0.65898876, | |
| "epoch": 1.5176151761517616, | |
| "grad_norm": 1.4296875, | |
| "learning_rate": 9.800285306704707e-05, | |
| "loss": 1.26696377, | |
| "memory(GiB)": 16.34, | |
| "step": 560, | |
| "train_speed(iter/s)": 0.14119 | |
| }, | |
| { | |
| "acc": 0.63674688, | |
| "epoch": 1.5311653116531165, | |
| "grad_norm": 1.5390625, | |
| "learning_rate": 9.746790299572039e-05, | |
| "loss": 1.38277016, | |
| "memory(GiB)": 16.34, | |
| "step": 565, | |
| "train_speed(iter/s)": 0.141349 | |
| }, | |
| { | |
| "acc": 0.67204466, | |
| "epoch": 1.5447154471544715, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 9.693295292439372e-05, | |
| "loss": 1.25243311, | |
| "memory(GiB)": 16.34, | |
| "step": 570, | |
| "train_speed(iter/s)": 0.141502 | |
| }, | |
| { | |
| "acc": 0.67878027, | |
| "epoch": 1.5582655826558267, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 9.639800285306703e-05, | |
| "loss": 1.18253031, | |
| "memory(GiB)": 16.34, | |
| "step": 575, | |
| "train_speed(iter/s)": 0.141663 | |
| }, | |
| { | |
| "acc": 0.64645357, | |
| "epoch": 1.5718157181571817, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 9.586305278174036e-05, | |
| "loss": 1.39789114, | |
| "memory(GiB)": 16.34, | |
| "step": 580, | |
| "train_speed(iter/s)": 0.141812 | |
| }, | |
| { | |
| "acc": 0.63896599, | |
| "epoch": 1.5853658536585367, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 9.532810271041368e-05, | |
| "loss": 1.37434454, | |
| "memory(GiB)": 16.34, | |
| "step": 585, | |
| "train_speed(iter/s)": 0.141961 | |
| }, | |
| { | |
| "acc": 0.67930498, | |
| "epoch": 1.5989159891598916, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 9.479315263908701e-05, | |
| "loss": 1.24675446, | |
| "memory(GiB)": 16.34, | |
| "step": 590, | |
| "train_speed(iter/s)": 0.142113 | |
| }, | |
| { | |
| "acc": 0.66582651, | |
| "epoch": 1.6124661246612466, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 9.425820256776034e-05, | |
| "loss": 1.33937092, | |
| "memory(GiB)": 16.34, | |
| "step": 595, | |
| "train_speed(iter/s)": 0.142256 | |
| }, | |
| { | |
| "acc": 0.65553112, | |
| "epoch": 1.6260162601626016, | |
| "grad_norm": 1.5859375, | |
| "learning_rate": 9.372325249643366e-05, | |
| "loss": 1.25127707, | |
| "memory(GiB)": 16.34, | |
| "step": 600, | |
| "train_speed(iter/s)": 0.142394 | |
| }, | |
| { | |
| "epoch": 1.6260162601626016, | |
| "eval_acc": 0.6233744299949333, | |
| "eval_loss": 1.5426762104034424, | |
| "eval_runtime": 44.3564, | |
| "eval_samples_per_second": 0.857, | |
| "eval_steps_per_second": 0.857, | |
| "step": 600 | |
| }, | |
| { | |
| "acc": 0.64805899, | |
| "epoch": 1.6395663956639566, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 9.318830242510699e-05, | |
| "loss": 1.30654621, | |
| "memory(GiB)": 16.34, | |
| "step": 605, | |
| "train_speed(iter/s)": 0.141073 | |
| }, | |
| { | |
| "acc": 0.67441335, | |
| "epoch": 1.6531165311653115, | |
| "grad_norm": 1.34375, | |
| "learning_rate": 9.26533523537803e-05, | |
| "loss": 1.20785751, | |
| "memory(GiB)": 16.34, | |
| "step": 610, | |
| "train_speed(iter/s)": 0.141227 | |
| }, | |
| { | |
| "acc": 0.64989614, | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 2.25, | |
| "learning_rate": 9.211840228245362e-05, | |
| "loss": 1.41231976, | |
| "memory(GiB)": 16.34, | |
| "step": 615, | |
| "train_speed(iter/s)": 0.141374 | |
| }, | |
| { | |
| "acc": 0.65118213, | |
| "epoch": 1.6802168021680217, | |
| "grad_norm": 1.34375, | |
| "learning_rate": 9.158345221112695e-05, | |
| "loss": 1.33156195, | |
| "memory(GiB)": 16.34, | |
| "step": 620, | |
| "train_speed(iter/s)": 0.141527 | |
| }, | |
| { | |
| "acc": 0.68491917, | |
| "epoch": 1.6937669376693767, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 9.104850213980028e-05, | |
| "loss": 1.15608921, | |
| "memory(GiB)": 16.34, | |
| "step": 625, | |
| "train_speed(iter/s)": 0.141656 | |
| }, | |
| { | |
| "acc": 0.6636075, | |
| "epoch": 1.7073170731707317, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 9.051355206847361e-05, | |
| "loss": 1.3377409, | |
| "memory(GiB)": 16.34, | |
| "step": 630, | |
| "train_speed(iter/s)": 0.141795 | |
| }, | |
| { | |
| "acc": 0.66115265, | |
| "epoch": 1.7208672086720869, | |
| "grad_norm": 1.796875, | |
| "learning_rate": 8.997860199714693e-05, | |
| "loss": 1.41666918, | |
| "memory(GiB)": 16.34, | |
| "step": 635, | |
| "train_speed(iter/s)": 0.141928 | |
| }, | |
| { | |
| "acc": 0.65316691, | |
| "epoch": 1.7344173441734418, | |
| "grad_norm": 1.875, | |
| "learning_rate": 8.944365192582024e-05, | |
| "loss": 1.37885714, | |
| "memory(GiB)": 16.34, | |
| "step": 640, | |
| "train_speed(iter/s)": 0.14206 | |
| }, | |
| { | |
| "acc": 0.63365035, | |
| "epoch": 1.7479674796747968, | |
| "grad_norm": 2.03125, | |
| "learning_rate": 8.890870185449357e-05, | |
| "loss": 1.43597651, | |
| "memory(GiB)": 16.34, | |
| "step": 645, | |
| "train_speed(iter/s)": 0.142195 | |
| }, | |
| { | |
| "acc": 0.6461009, | |
| "epoch": 1.7615176151761518, | |
| "grad_norm": 1.390625, | |
| "learning_rate": 8.837375178316689e-05, | |
| "loss": 1.40720987, | |
| "memory(GiB)": 16.34, | |
| "step": 650, | |
| "train_speed(iter/s)": 0.142326 | |
| }, | |
| { | |
| "epoch": 1.7615176151761518, | |
| "eval_acc": 0.6223610876541125, | |
| "eval_loss": 1.5279418230056763, | |
| "eval_runtime": 44.3195, | |
| "eval_samples_per_second": 0.857, | |
| "eval_steps_per_second": 0.857, | |
| "step": 650 | |
| }, | |
| { | |
| "acc": 0.68952079, | |
| "epoch": 1.7750677506775068, | |
| "grad_norm": 1.5859375, | |
| "learning_rate": 8.783880171184023e-05, | |
| "loss": 1.1804883, | |
| "memory(GiB)": 16.34, | |
| "step": 655, | |
| "train_speed(iter/s)": 0.141093 | |
| }, | |
| { | |
| "acc": 0.63278737, | |
| "epoch": 1.7886178861788617, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 8.730385164051355e-05, | |
| "loss": 1.45864544, | |
| "memory(GiB)": 16.34, | |
| "step": 660, | |
| "train_speed(iter/s)": 0.141227 | |
| }, | |
| { | |
| "acc": 0.6499536, | |
| "epoch": 1.8021680216802167, | |
| "grad_norm": 1.0859375, | |
| "learning_rate": 8.676890156918687e-05, | |
| "loss": 1.36629667, | |
| "memory(GiB)": 16.34, | |
| "step": 665, | |
| "train_speed(iter/s)": 0.14135 | |
| }, | |
| { | |
| "acc": 0.66262636, | |
| "epoch": 1.8157181571815717, | |
| "grad_norm": 2.8125, | |
| "learning_rate": 8.62339514978602e-05, | |
| "loss": 1.28030796, | |
| "memory(GiB)": 16.34, | |
| "step": 670, | |
| "train_speed(iter/s)": 0.141486 | |
| }, | |
| { | |
| "acc": 0.6478013, | |
| "epoch": 1.8292682926829267, | |
| "grad_norm": 1.3203125, | |
| "learning_rate": 8.569900142653351e-05, | |
| "loss": 1.37910089, | |
| "memory(GiB)": 16.34, | |
| "step": 675, | |
| "train_speed(iter/s)": 0.141617 | |
| }, | |
| { | |
| "acc": 0.65061078, | |
| "epoch": 1.8428184281842819, | |
| "grad_norm": 1.859375, | |
| "learning_rate": 8.516405135520683e-05, | |
| "loss": 1.24110394, | |
| "memory(GiB)": 16.34, | |
| "step": 680, | |
| "train_speed(iter/s)": 0.141749 | |
| }, | |
| { | |
| "acc": 0.66720443, | |
| "epoch": 1.8563685636856369, | |
| "grad_norm": 1.2734375, | |
| "learning_rate": 8.462910128388016e-05, | |
| "loss": 1.36949673, | |
| "memory(GiB)": 16.34, | |
| "step": 685, | |
| "train_speed(iter/s)": 0.141882 | |
| }, | |
| { | |
| "acc": 0.65051932, | |
| "epoch": 1.8699186991869918, | |
| "grad_norm": 1.796875, | |
| "learning_rate": 8.409415121255349e-05, | |
| "loss": 1.3470686, | |
| "memory(GiB)": 16.34, | |
| "step": 690, | |
| "train_speed(iter/s)": 0.142007 | |
| }, | |
| { | |
| "acc": 0.64647999, | |
| "epoch": 1.883468834688347, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 8.355920114122682e-05, | |
| "loss": 1.27561255, | |
| "memory(GiB)": 16.34, | |
| "step": 695, | |
| "train_speed(iter/s)": 0.142124 | |
| }, | |
| { | |
| "acc": 0.64771528, | |
| "epoch": 1.897018970189702, | |
| "grad_norm": 2.5, | |
| "learning_rate": 8.302425106990014e-05, | |
| "loss": 1.3874403, | |
| "memory(GiB)": 16.34, | |
| "step": 700, | |
| "train_speed(iter/s)": 0.142242 | |
| }, | |
| { | |
| "epoch": 1.897018970189702, | |
| "eval_acc": 0.6274277993582165, | |
| "eval_loss": 1.5236802101135254, | |
| "eval_runtime": 44.4112, | |
| "eval_samples_per_second": 0.856, | |
| "eval_steps_per_second": 0.856, | |
| "step": 700 | |
| }, | |
| { | |
| "acc": 0.65817127, | |
| "epoch": 1.910569105691057, | |
| "grad_norm": 1.3203125, | |
| "learning_rate": 8.248930099857345e-05, | |
| "loss": 1.29632025, | |
| "memory(GiB)": 16.34, | |
| "step": 705, | |
| "train_speed(iter/s)": 0.141099 | |
| }, | |
| { | |
| "acc": 0.64647436, | |
| "epoch": 1.924119241192412, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 8.195435092724678e-05, | |
| "loss": 1.39382238, | |
| "memory(GiB)": 16.34, | |
| "step": 710, | |
| "train_speed(iter/s)": 0.141217 | |
| }, | |
| { | |
| "acc": 0.65741391, | |
| "epoch": 1.937669376693767, | |
| "grad_norm": 0.98828125, | |
| "learning_rate": 8.14194008559201e-05, | |
| "loss": 1.32606802, | |
| "memory(GiB)": 16.34, | |
| "step": 715, | |
| "train_speed(iter/s)": 0.141337 | |
| }, | |
| { | |
| "acc": 0.65078535, | |
| "epoch": 1.951219512195122, | |
| "grad_norm": 1.6171875, | |
| "learning_rate": 8.088445078459343e-05, | |
| "loss": 1.28092451, | |
| "memory(GiB)": 16.34, | |
| "step": 720, | |
| "train_speed(iter/s)": 0.141457 | |
| }, | |
| { | |
| "acc": 0.64983764, | |
| "epoch": 1.9647696476964769, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 8.034950071326676e-05, | |
| "loss": 1.35295801, | |
| "memory(GiB)": 16.34, | |
| "step": 725, | |
| "train_speed(iter/s)": 0.141585 | |
| }, | |
| { | |
| "acc": 0.64880919, | |
| "epoch": 1.9783197831978319, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 7.981455064194009e-05, | |
| "loss": 1.38945732, | |
| "memory(GiB)": 16.34, | |
| "step": 730, | |
| "train_speed(iter/s)": 0.141703 | |
| }, | |
| { | |
| "acc": 0.64617214, | |
| "epoch": 1.9918699186991868, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 7.927960057061341e-05, | |
| "loss": 1.31189098, | |
| "memory(GiB)": 16.34, | |
| "step": 735, | |
| "train_speed(iter/s)": 0.141817 | |
| }, | |
| { | |
| "acc": 0.67255039, | |
| "epoch": 2.005420054200542, | |
| "grad_norm": 1.3359375, | |
| "learning_rate": 7.874465049928672e-05, | |
| "loss": 1.17746153, | |
| "memory(GiB)": 16.34, | |
| "step": 740, | |
| "train_speed(iter/s)": 0.141669 | |
| }, | |
| { | |
| "acc": 0.69897633, | |
| "epoch": 2.0189701897018972, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 7.820970042796005e-05, | |
| "loss": 1.03704672, | |
| "memory(GiB)": 16.34, | |
| "step": 745, | |
| "train_speed(iter/s)": 0.141785 | |
| }, | |
| { | |
| "acc": 0.71377811, | |
| "epoch": 2.032520325203252, | |
| "grad_norm": 1.640625, | |
| "learning_rate": 7.767475035663337e-05, | |
| "loss": 1.04516726, | |
| "memory(GiB)": 16.34, | |
| "step": 750, | |
| "train_speed(iter/s)": 0.141905 | |
| }, | |
| { | |
| "epoch": 2.032520325203252, | |
| "eval_acc": 0.6208410741428813, | |
| "eval_loss": 1.603255271911621, | |
| "eval_runtime": 44.4359, | |
| "eval_samples_per_second": 0.855, | |
| "eval_steps_per_second": 0.855, | |
| "step": 750 | |
| }, | |
| { | |
| "acc": 0.73705945, | |
| "epoch": 2.046070460704607, | |
| "grad_norm": 1.8203125, | |
| "learning_rate": 7.713980028530669e-05, | |
| "loss": 0.9383255, | |
| "memory(GiB)": 16.34, | |
| "step": 755, | |
| "train_speed(iter/s)": 0.14084 | |
| }, | |
| { | |
| "acc": 0.73054934, | |
| "epoch": 2.059620596205962, | |
| "grad_norm": 1.8203125, | |
| "learning_rate": 7.660485021398003e-05, | |
| "loss": 0.91379232, | |
| "memory(GiB)": 16.34, | |
| "step": 760, | |
| "train_speed(iter/s)": 0.140958 | |
| }, | |
| { | |
| "acc": 0.73154573, | |
| "epoch": 2.073170731707317, | |
| "grad_norm": 1.6796875, | |
| "learning_rate": 7.606990014265335e-05, | |
| "loss": 0.97700481, | |
| "memory(GiB)": 16.34, | |
| "step": 765, | |
| "train_speed(iter/s)": 0.141076 | |
| }, | |
| { | |
| "acc": 0.72998781, | |
| "epoch": 2.086720867208672, | |
| "grad_norm": 1.4140625, | |
| "learning_rate": 7.553495007132668e-05, | |
| "loss": 0.93853807, | |
| "memory(GiB)": 16.34, | |
| "step": 770, | |
| "train_speed(iter/s)": 0.141186 | |
| }, | |
| { | |
| "acc": 0.73213534, | |
| "epoch": 2.100271002710027, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 7.5e-05, | |
| "loss": 0.94589176, | |
| "memory(GiB)": 16.34, | |
| "step": 775, | |
| "train_speed(iter/s)": 0.141303 | |
| }, | |
| { | |
| "acc": 0.73035493, | |
| "epoch": 2.113821138211382, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 7.446504992867331e-05, | |
| "loss": 0.95597754, | |
| "memory(GiB)": 16.34, | |
| "step": 780, | |
| "train_speed(iter/s)": 0.141422 | |
| }, | |
| { | |
| "acc": 0.73428354, | |
| "epoch": 2.127371273712737, | |
| "grad_norm": 1.484375, | |
| "learning_rate": 7.393009985734664e-05, | |
| "loss": 0.98254423, | |
| "memory(GiB)": 16.34, | |
| "step": 785, | |
| "train_speed(iter/s)": 0.141543 | |
| }, | |
| { | |
| "acc": 0.69750729, | |
| "epoch": 2.140921409214092, | |
| "grad_norm": 1.828125, | |
| "learning_rate": 7.339514978601997e-05, | |
| "loss": 1.10206041, | |
| "memory(GiB)": 16.34, | |
| "step": 790, | |
| "train_speed(iter/s)": 0.141651 | |
| }, | |
| { | |
| "acc": 0.73764381, | |
| "epoch": 2.154471544715447, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 7.286019971469329e-05, | |
| "loss": 0.84075432, | |
| "memory(GiB)": 16.34, | |
| "step": 795, | |
| "train_speed(iter/s)": 0.141765 | |
| }, | |
| { | |
| "acc": 0.72124152, | |
| "epoch": 2.168021680216802, | |
| "grad_norm": 1.4375, | |
| "learning_rate": 7.23252496433666e-05, | |
| "loss": 1.05863771, | |
| "memory(GiB)": 16.34, | |
| "step": 800, | |
| "train_speed(iter/s)": 0.141874 | |
| }, | |
| { | |
| "epoch": 2.168021680216802, | |
| "eval_acc": 0.6240499915554805, | |
| "eval_loss": 1.61227285861969, | |
| "eval_runtime": 44.3145, | |
| "eval_samples_per_second": 0.858, | |
| "eval_steps_per_second": 0.858, | |
| "step": 800 | |
| }, | |
| { | |
| "acc": 0.73518519, | |
| "epoch": 2.181571815718157, | |
| "grad_norm": 1.546875, | |
| "learning_rate": 7.179029957203993e-05, | |
| "loss": 0.96204119, | |
| "memory(GiB)": 16.34, | |
| "step": 805, | |
| "train_speed(iter/s)": 0.140881 | |
| }, | |
| { | |
| "acc": 0.74290161, | |
| "epoch": 2.1951219512195124, | |
| "grad_norm": 2.015625, | |
| "learning_rate": 7.125534950071326e-05, | |
| "loss": 0.92387733, | |
| "memory(GiB)": 16.34, | |
| "step": 810, | |
| "train_speed(iter/s)": 0.14021 | |
| }, | |
| { | |
| "acc": 0.73101602, | |
| "epoch": 2.2086720867208673, | |
| "grad_norm": 1.6015625, | |
| "learning_rate": 7.072039942938658e-05, | |
| "loss": 0.96248655, | |
| "memory(GiB)": 16.34, | |
| "step": 815, | |
| "train_speed(iter/s)": 0.140323 | |
| }, | |
| { | |
| "acc": 0.73887796, | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 1.796875, | |
| "learning_rate": 7.018544935805991e-05, | |
| "loss": 0.94751673, | |
| "memory(GiB)": 16.34, | |
| "step": 820, | |
| "train_speed(iter/s)": 0.140441 | |
| }, | |
| { | |
| "acc": 0.75793715, | |
| "epoch": 2.2357723577235773, | |
| "grad_norm": 2.015625, | |
| "learning_rate": 6.965049928673323e-05, | |
| "loss": 0.84263477, | |
| "memory(GiB)": 16.34, | |
| "step": 825, | |
| "train_speed(iter/s)": 0.140555 | |
| }, | |
| { | |
| "acc": 0.75873909, | |
| "epoch": 2.2493224932249323, | |
| "grad_norm": 1.953125, | |
| "learning_rate": 6.911554921540656e-05, | |
| "loss": 0.89048252, | |
| "memory(GiB)": 16.34, | |
| "step": 830, | |
| "train_speed(iter/s)": 0.14067 | |
| }, | |
| { | |
| "acc": 0.74456077, | |
| "epoch": 2.2628726287262872, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 6.858059914407987e-05, | |
| "loss": 0.90777779, | |
| "memory(GiB)": 16.34, | |
| "step": 835, | |
| "train_speed(iter/s)": 0.140775 | |
| }, | |
| { | |
| "acc": 0.75809846, | |
| "epoch": 2.2764227642276422, | |
| "grad_norm": 2.0625, | |
| "learning_rate": 6.80456490727532e-05, | |
| "loss": 0.87556753, | |
| "memory(GiB)": 16.34, | |
| "step": 840, | |
| "train_speed(iter/s)": 0.14088 | |
| }, | |
| { | |
| "acc": 0.73194971, | |
| "epoch": 2.289972899728997, | |
| "grad_norm": 2.765625, | |
| "learning_rate": 6.751069900142653e-05, | |
| "loss": 0.9232769, | |
| "memory(GiB)": 16.34, | |
| "step": 845, | |
| "train_speed(iter/s)": 0.140986 | |
| }, | |
| { | |
| "acc": 0.74470835, | |
| "epoch": 2.303523035230352, | |
| "grad_norm": 1.8515625, | |
| "learning_rate": 6.697574893009985e-05, | |
| "loss": 0.88600664, | |
| "memory(GiB)": 16.34, | |
| "step": 850, | |
| "train_speed(iter/s)": 0.141086 | |
| }, | |
| { | |
| "epoch": 2.303523035230352, | |
| "eval_acc": 0.6174632663401453, | |
| "eval_loss": 1.660492181777954, | |
| "eval_runtime": 44.1393, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 850 | |
| }, | |
| { | |
| "acc": 0.76063514, | |
| "epoch": 2.317073170731707, | |
| "grad_norm": 2.40625, | |
| "learning_rate": 6.644079885877318e-05, | |
| "loss": 0.80236473, | |
| "memory(GiB)": 16.34, | |
| "step": 855, | |
| "train_speed(iter/s)": 0.140168 | |
| }, | |
| { | |
| "acc": 0.71505499, | |
| "epoch": 2.330623306233062, | |
| "grad_norm": 2.671875, | |
| "learning_rate": 6.59058487874465e-05, | |
| "loss": 1.01317081, | |
| "memory(GiB)": 16.34, | |
| "step": 860, | |
| "train_speed(iter/s)": 0.140281 | |
| }, | |
| { | |
| "acc": 0.73396034, | |
| "epoch": 2.3441734417344176, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 6.537089871611983e-05, | |
| "loss": 0.96496754, | |
| "memory(GiB)": 16.34, | |
| "step": 865, | |
| "train_speed(iter/s)": 0.140387 | |
| }, | |
| { | |
| "acc": 0.74145699, | |
| "epoch": 2.3577235772357725, | |
| "grad_norm": 1.9453125, | |
| "learning_rate": 6.483594864479316e-05, | |
| "loss": 1.00490999, | |
| "memory(GiB)": 16.34, | |
| "step": 870, | |
| "train_speed(iter/s)": 0.140492 | |
| }, | |
| { | |
| "acc": 0.76035104, | |
| "epoch": 2.3712737127371275, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 6.430099857346647e-05, | |
| "loss": 0.93969469, | |
| "memory(GiB)": 16.34, | |
| "step": 875, | |
| "train_speed(iter/s)": 0.140599 | |
| }, | |
| { | |
| "acc": 0.75214877, | |
| "epoch": 2.3848238482384825, | |
| "grad_norm": 1.9296875, | |
| "learning_rate": 6.376604850213979e-05, | |
| "loss": 0.9385232, | |
| "memory(GiB)": 16.34, | |
| "step": 880, | |
| "train_speed(iter/s)": 0.140705 | |
| }, | |
| { | |
| "acc": 0.73846526, | |
| "epoch": 2.3983739837398375, | |
| "grad_norm": 2.421875, | |
| "learning_rate": 6.323109843081312e-05, | |
| "loss": 0.95887814, | |
| "memory(GiB)": 16.34, | |
| "step": 885, | |
| "train_speed(iter/s)": 0.140812 | |
| }, | |
| { | |
| "acc": 0.72576594, | |
| "epoch": 2.4119241192411924, | |
| "grad_norm": 1.46875, | |
| "learning_rate": 6.269614835948645e-05, | |
| "loss": 0.95252094, | |
| "memory(GiB)": 16.34, | |
| "step": 890, | |
| "train_speed(iter/s)": 0.140912 | |
| }, | |
| { | |
| "acc": 0.74866586, | |
| "epoch": 2.4254742547425474, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 6.216119828815977e-05, | |
| "loss": 0.91533632, | |
| "memory(GiB)": 16.34, | |
| "step": 895, | |
| "train_speed(iter/s)": 0.141011 | |
| }, | |
| { | |
| "acc": 0.72289066, | |
| "epoch": 2.4390243902439024, | |
| "grad_norm": 1.375, | |
| "learning_rate": 6.162624821683308e-05, | |
| "loss": 0.996418, | |
| "memory(GiB)": 16.34, | |
| "step": 900, | |
| "train_speed(iter/s)": 0.13966 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "eval_acc": 0.6152676912683668, | |
| "eval_loss": 1.6503233909606934, | |
| "eval_runtime": 44.1939, | |
| "eval_samples_per_second": 0.86, | |
| "eval_steps_per_second": 0.86, | |
| "step": 900 | |
| }, | |
| { | |
| "acc": 0.73535914, | |
| "epoch": 2.4525745257452574, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 6.109129814550641e-05, | |
| "loss": 1.00612879, | |
| "memory(GiB)": 16.34, | |
| "step": 905, | |
| "train_speed(iter/s)": 0.138805 | |
| }, | |
| { | |
| "acc": 0.7277998, | |
| "epoch": 2.4661246612466123, | |
| "grad_norm": 2.328125, | |
| "learning_rate": 6.0556348074179737e-05, | |
| "loss": 1.00636635, | |
| "memory(GiB)": 16.34, | |
| "step": 910, | |
| "train_speed(iter/s)": 0.138904 | |
| }, | |
| { | |
| "acc": 0.70795035, | |
| "epoch": 2.4796747967479673, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 6.002139800285306e-05, | |
| "loss": 0.98440151, | |
| "memory(GiB)": 16.34, | |
| "step": 915, | |
| "train_speed(iter/s)": 0.139013 | |
| }, | |
| { | |
| "acc": 0.7650095, | |
| "epoch": 2.4932249322493227, | |
| "grad_norm": 2.25, | |
| "learning_rate": 5.948644793152638e-05, | |
| "loss": 0.77239523, | |
| "memory(GiB)": 16.34, | |
| "step": 920, | |
| "train_speed(iter/s)": 0.139126 | |
| }, | |
| { | |
| "acc": 0.71504927, | |
| "epoch": 2.5067750677506773, | |
| "grad_norm": 1.6328125, | |
| "learning_rate": 5.895149786019971e-05, | |
| "loss": 1.01518288, | |
| "memory(GiB)": 16.34, | |
| "step": 925, | |
| "train_speed(iter/s)": 0.139233 | |
| }, | |
| { | |
| "acc": 0.71582847, | |
| "epoch": 2.5203252032520327, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 5.841654778887303e-05, | |
| "loss": 0.97862291, | |
| "memory(GiB)": 16.34, | |
| "step": 930, | |
| "train_speed(iter/s)": 0.13934 | |
| }, | |
| { | |
| "acc": 0.74984369, | |
| "epoch": 2.5338753387533877, | |
| "grad_norm": 1.8515625, | |
| "learning_rate": 5.788159771754635e-05, | |
| "loss": 0.85021992, | |
| "memory(GiB)": 16.34, | |
| "step": 935, | |
| "train_speed(iter/s)": 0.139448 | |
| }, | |
| { | |
| "acc": 0.74142261, | |
| "epoch": 2.5474254742547426, | |
| "grad_norm": 1.796875, | |
| "learning_rate": 5.734664764621968e-05, | |
| "loss": 0.91843338, | |
| "memory(GiB)": 16.34, | |
| "step": 940, | |
| "train_speed(iter/s)": 0.139564 | |
| }, | |
| { | |
| "acc": 0.71266222, | |
| "epoch": 2.5609756097560976, | |
| "grad_norm": 1.875, | |
| "learning_rate": 5.6811697574893007e-05, | |
| "loss": 1.14017658, | |
| "memory(GiB)": 16.34, | |
| "step": 945, | |
| "train_speed(iter/s)": 0.139667 | |
| }, | |
| { | |
| "acc": 0.73374839, | |
| "epoch": 2.5745257452574526, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 5.627674750356633e-05, | |
| "loss": 0.99979248, | |
| "memory(GiB)": 16.34, | |
| "step": 950, | |
| "train_speed(iter/s)": 0.139769 | |
| }, | |
| { | |
| "epoch": 2.5745257452574526, | |
| "eval_acc": 0.6161121432190508, | |
| "eval_loss": 1.6511973142623901, | |
| "eval_runtime": 44.1881, | |
| "eval_samples_per_second": 0.86, | |
| "eval_steps_per_second": 0.86, | |
| "step": 950 | |
| }, | |
| { | |
| "acc": 0.74381332, | |
| "epoch": 2.5880758807588076, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 5.5741797432239646e-05, | |
| "loss": 0.93189411, | |
| "memory(GiB)": 16.34, | |
| "step": 955, | |
| "train_speed(iter/s)": 0.138972 | |
| }, | |
| { | |
| "acc": 0.73305674, | |
| "epoch": 2.6016260162601625, | |
| "grad_norm": 2.4375, | |
| "learning_rate": 5.5206847360912977e-05, | |
| "loss": 0.93946905, | |
| "memory(GiB)": 16.34, | |
| "step": 960, | |
| "train_speed(iter/s)": 0.139079 | |
| }, | |
| { | |
| "acc": 0.72961435, | |
| "epoch": 2.6151761517615175, | |
| "grad_norm": 1.84375, | |
| "learning_rate": 5.46718972895863e-05, | |
| "loss": 0.99850683, | |
| "memory(GiB)": 16.34, | |
| "step": 965, | |
| "train_speed(iter/s)": 0.139175 | |
| }, | |
| { | |
| "acc": 0.7133184, | |
| "epoch": 2.6287262872628725, | |
| "grad_norm": 1.96875, | |
| "learning_rate": 5.413694721825962e-05, | |
| "loss": 1.03378878, | |
| "memory(GiB)": 16.34, | |
| "step": 970, | |
| "train_speed(iter/s)": 0.139282 | |
| }, | |
| { | |
| "acc": 0.71164918, | |
| "epoch": 2.642276422764228, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 5.360199714693295e-05, | |
| "loss": 1.02268944, | |
| "memory(GiB)": 16.34, | |
| "step": 975, | |
| "train_speed(iter/s)": 0.139382 | |
| }, | |
| { | |
| "acc": 0.71824646, | |
| "epoch": 2.6558265582655824, | |
| "grad_norm": 1.84375, | |
| "learning_rate": 5.306704707560627e-05, | |
| "loss": 1.06014738, | |
| "memory(GiB)": 16.34, | |
| "step": 980, | |
| "train_speed(iter/s)": 0.139479 | |
| }, | |
| { | |
| "acc": 0.75523286, | |
| "epoch": 2.669376693766938, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 5.253209700427959e-05, | |
| "loss": 0.80533791, | |
| "memory(GiB)": 16.34, | |
| "step": 985, | |
| "train_speed(iter/s)": 0.139581 | |
| }, | |
| { | |
| "acc": 0.72265592, | |
| "epoch": 2.682926829268293, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 5.199714693295292e-05, | |
| "loss": 1.0223958, | |
| "memory(GiB)": 16.34, | |
| "step": 990, | |
| "train_speed(iter/s)": 0.139675 | |
| }, | |
| { | |
| "acc": 0.74640193, | |
| "epoch": 2.696476964769648, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 5.1462196861626247e-05, | |
| "loss": 0.8896265, | |
| "memory(GiB)": 16.34, | |
| "step": 995, | |
| "train_speed(iter/s)": 0.139767 | |
| }, | |
| { | |
| "acc": 0.75235152, | |
| "epoch": 2.710027100271003, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 5.092724679029957e-05, | |
| "loss": 0.87937946, | |
| "memory(GiB)": 16.34, | |
| "step": 1000, | |
| "train_speed(iter/s)": 0.139867 | |
| }, | |
| { | |
| "epoch": 2.710027100271003, | |
| "eval_acc": 0.6123965546360413, | |
| "eval_loss": 1.6541699171066284, | |
| "eval_runtime": 44.2136, | |
| "eval_samples_per_second": 0.859, | |
| "eval_steps_per_second": 0.859, | |
| "step": 1000 | |
| }, | |
| { | |
| "acc": 0.73264089, | |
| "epoch": 2.7235772357723578, | |
| "grad_norm": 1.96875, | |
| "learning_rate": 5.0392296718972886e-05, | |
| "loss": 0.95834293, | |
| "memory(GiB)": 16.34, | |
| "step": 1005, | |
| "train_speed(iter/s)": 0.139102 | |
| }, | |
| { | |
| "acc": 0.76931543, | |
| "epoch": 2.7371273712737128, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 4.9857346647646217e-05, | |
| "loss": 0.82889891, | |
| "memory(GiB)": 16.34, | |
| "step": 1010, | |
| "train_speed(iter/s)": 0.139199 | |
| }, | |
| { | |
| "acc": 0.74891815, | |
| "epoch": 2.7506775067750677, | |
| "grad_norm": 2.65625, | |
| "learning_rate": 4.932239657631954e-05, | |
| "loss": 0.93503094, | |
| "memory(GiB)": 16.34, | |
| "step": 1015, | |
| "train_speed(iter/s)": 0.139298 | |
| }, | |
| { | |
| "acc": 0.72551074, | |
| "epoch": 2.7642276422764227, | |
| "grad_norm": 1.5703125, | |
| "learning_rate": 4.878744650499286e-05, | |
| "loss": 1.06086788, | |
| "memory(GiB)": 16.34, | |
| "step": 1020, | |
| "train_speed(iter/s)": 0.139383 | |
| }, | |
| { | |
| "acc": 0.75321589, | |
| "epoch": 2.7777777777777777, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 4.825249643366619e-05, | |
| "loss": 0.83553734, | |
| "memory(GiB)": 16.34, | |
| "step": 1025, | |
| "train_speed(iter/s)": 0.139478 | |
| }, | |
| { | |
| "acc": 0.73832054, | |
| "epoch": 2.7913279132791327, | |
| "grad_norm": 1.53125, | |
| "learning_rate": 4.771754636233951e-05, | |
| "loss": 0.94093418, | |
| "memory(GiB)": 16.34, | |
| "step": 1030, | |
| "train_speed(iter/s)": 0.139572 | |
| }, | |
| { | |
| "acc": 0.76351671, | |
| "epoch": 2.8048780487804876, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 4.718259629101283e-05, | |
| "loss": 0.84423561, | |
| "memory(GiB)": 16.34, | |
| "step": 1035, | |
| "train_speed(iter/s)": 0.139664 | |
| }, | |
| { | |
| "acc": 0.75100279, | |
| "epoch": 2.818428184281843, | |
| "grad_norm": 1.890625, | |
| "learning_rate": 4.6647646219686156e-05, | |
| "loss": 0.91757078, | |
| "memory(GiB)": 16.34, | |
| "step": 1040, | |
| "train_speed(iter/s)": 0.139754 | |
| }, | |
| { | |
| "acc": 0.73302913, | |
| "epoch": 2.8319783197831976, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 4.6112696148359487e-05, | |
| "loss": 0.94331446, | |
| "memory(GiB)": 16.34, | |
| "step": 1045, | |
| "train_speed(iter/s)": 0.139841 | |
| }, | |
| { | |
| "acc": 0.73944206, | |
| "epoch": 2.845528455284553, | |
| "grad_norm": 1.59375, | |
| "learning_rate": 4.55777460770328e-05, | |
| "loss": 0.98726377, | |
| "memory(GiB)": 16.34, | |
| "step": 1050, | |
| "train_speed(iter/s)": 0.139931 | |
| }, | |
| { | |
| "epoch": 2.845528455284553, | |
| "eval_acc": 0.6145921297078196, | |
| "eval_loss": 1.6571751832962036, | |
| "eval_runtime": 44.3705, | |
| "eval_samples_per_second": 0.856, | |
| "eval_steps_per_second": 0.856, | |
| "step": 1050 | |
| }, | |
| { | |
| "acc": 0.73031135, | |
| "epoch": 2.859078590785908, | |
| "grad_norm": 2.25, | |
| "learning_rate": 4.5042796005706126e-05, | |
| "loss": 0.95681438, | |
| "memory(GiB)": 16.34, | |
| "step": 1055, | |
| "train_speed(iter/s)": 0.139201 | |
| }, | |
| { | |
| "acc": 0.71694627, | |
| "epoch": 2.872628726287263, | |
| "grad_norm": 1.6328125, | |
| "learning_rate": 4.4507845934379456e-05, | |
| "loss": 1.02266083, | |
| "memory(GiB)": 16.34, | |
| "step": 1060, | |
| "train_speed(iter/s)": 0.139294 | |
| }, | |
| { | |
| "acc": 0.72791638, | |
| "epoch": 2.886178861788618, | |
| "grad_norm": 1.7578125, | |
| "learning_rate": 4.397289586305278e-05, | |
| "loss": 0.99341927, | |
| "memory(GiB)": 16.34, | |
| "step": 1065, | |
| "train_speed(iter/s)": 0.139385 | |
| }, | |
| { | |
| "acc": 0.74939594, | |
| "epoch": 2.899728997289973, | |
| "grad_norm": 1.921875, | |
| "learning_rate": 4.34379457917261e-05, | |
| "loss": 0.91077061, | |
| "memory(GiB)": 16.34, | |
| "step": 1070, | |
| "train_speed(iter/s)": 0.139478 | |
| }, | |
| { | |
| "acc": 0.72694654, | |
| "epoch": 2.913279132791328, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 4.290299572039942e-05, | |
| "loss": 0.98774853, | |
| "memory(GiB)": 16.34, | |
| "step": 1075, | |
| "train_speed(iter/s)": 0.139564 | |
| }, | |
| { | |
| "acc": 0.70588508, | |
| "epoch": 2.926829268292683, | |
| "grad_norm": 2.15625, | |
| "learning_rate": 4.236804564907275e-05, | |
| "loss": 1.07887812, | |
| "memory(GiB)": 16.34, | |
| "step": 1080, | |
| "train_speed(iter/s)": 0.139657 | |
| }, | |
| { | |
| "acc": 0.74654303, | |
| "epoch": 2.940379403794038, | |
| "grad_norm": 1.7109375, | |
| "learning_rate": 4.183309557774607e-05, | |
| "loss": 0.91062069, | |
| "memory(GiB)": 16.34, | |
| "step": 1085, | |
| "train_speed(iter/s)": 0.139743 | |
| }, | |
| { | |
| "acc": 0.73493595, | |
| "epoch": 2.953929539295393, | |
| "grad_norm": 2.03125, | |
| "learning_rate": 4.1298145506419396e-05, | |
| "loss": 0.92819033, | |
| "memory(GiB)": 16.34, | |
| "step": 1090, | |
| "train_speed(iter/s)": 0.139829 | |
| }, | |
| { | |
| "acc": 0.71466756, | |
| "epoch": 2.9674796747967482, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 4.0763195435092727e-05, | |
| "loss": 1.01220913, | |
| "memory(GiB)": 16.34, | |
| "step": 1095, | |
| "train_speed(iter/s)": 0.139916 | |
| }, | |
| { | |
| "acc": 0.7607831, | |
| "epoch": 2.9810298102981028, | |
| "grad_norm": 2.375, | |
| "learning_rate": 4.022824536376604e-05, | |
| "loss": 0.8505785, | |
| "memory(GiB)": 16.34, | |
| "step": 1100, | |
| "train_speed(iter/s)": 0.140004 | |
| }, | |
| { | |
| "epoch": 2.9810298102981028, | |
| "eval_acc": 0.6206721837527445, | |
| "eval_loss": 1.6525288820266724, | |
| "eval_runtime": 44.2124, | |
| "eval_samples_per_second": 0.859, | |
| "eval_steps_per_second": 0.859, | |
| "step": 1100 | |
| }, | |
| { | |
| "acc": 0.74116454, | |
| "epoch": 2.994579945799458, | |
| "grad_norm": 2.65625, | |
| "learning_rate": 3.9693295292439366e-05, | |
| "loss": 0.92264805, | |
| "memory(GiB)": 16.34, | |
| "step": 1105, | |
| "train_speed(iter/s)": 0.139311 | |
| }, | |
| { | |
| "acc": 0.8182869, | |
| "epoch": 3.008130081300813, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 3.915834522111269e-05, | |
| "loss": 0.6927434, | |
| "memory(GiB)": 16.34, | |
| "step": 1110, | |
| "train_speed(iter/s)": 0.13923 | |
| }, | |
| { | |
| "acc": 0.83515587, | |
| "epoch": 3.021680216802168, | |
| "grad_norm": 1.9140625, | |
| "learning_rate": 3.862339514978602e-05, | |
| "loss": 0.58284421, | |
| "memory(GiB)": 16.34, | |
| "step": 1115, | |
| "train_speed(iter/s)": 0.139319 | |
| }, | |
| { | |
| "acc": 0.84008894, | |
| "epoch": 3.035230352303523, | |
| "grad_norm": 1.8828125, | |
| "learning_rate": 3.808844507845934e-05, | |
| "loss": 0.59128432, | |
| "memory(GiB)": 16.34, | |
| "step": 1120, | |
| "train_speed(iter/s)": 0.139405 | |
| }, | |
| { | |
| "acc": 0.84298267, | |
| "epoch": 3.048780487804878, | |
| "grad_norm": 2.4375, | |
| "learning_rate": 3.755349500713266e-05, | |
| "loss": 0.56685705, | |
| "memory(GiB)": 16.34, | |
| "step": 1125, | |
| "train_speed(iter/s)": 0.139493 | |
| }, | |
| { | |
| "acc": 0.83680372, | |
| "epoch": 3.062330623306233, | |
| "grad_norm": 3.171875, | |
| "learning_rate": 3.701854493580599e-05, | |
| "loss": 0.55599914, | |
| "memory(GiB)": 16.34, | |
| "step": 1130, | |
| "train_speed(iter/s)": 0.139571 | |
| }, | |
| { | |
| "acc": 0.81269236, | |
| "epoch": 3.075880758807588, | |
| "grad_norm": 2.40625, | |
| "learning_rate": 3.648359486447931e-05, | |
| "loss": 0.63413367, | |
| "memory(GiB)": 16.34, | |
| "step": 1135, | |
| "train_speed(iter/s)": 0.139659 | |
| }, | |
| { | |
| "acc": 0.8300642, | |
| "epoch": 3.089430894308943, | |
| "grad_norm": 2.71875, | |
| "learning_rate": 3.5948644793152636e-05, | |
| "loss": 0.59288554, | |
| "memory(GiB)": 16.34, | |
| "step": 1140, | |
| "train_speed(iter/s)": 0.139738 | |
| }, | |
| { | |
| "acc": 0.84225779, | |
| "epoch": 3.102981029810298, | |
| "grad_norm": 2.703125, | |
| "learning_rate": 3.541369472182596e-05, | |
| "loss": 0.57104263, | |
| "memory(GiB)": 16.34, | |
| "step": 1145, | |
| "train_speed(iter/s)": 0.13982 | |
| }, | |
| { | |
| "acc": 0.86128368, | |
| "epoch": 3.116531165311653, | |
| "grad_norm": 3.0, | |
| "learning_rate": 3.487874465049928e-05, | |
| "loss": 0.47776127, | |
| "memory(GiB)": 16.34, | |
| "step": 1150, | |
| "train_speed(iter/s)": 0.139903 | |
| }, | |
| { | |
| "epoch": 3.116531165311653, | |
| "eval_acc": 0.6091876372234419, | |
| "eval_loss": 1.9067459106445312, | |
| "eval_runtime": 44.213, | |
| "eval_samples_per_second": 0.859, | |
| "eval_steps_per_second": 0.859, | |
| "step": 1150 | |
| }, | |
| { | |
| "acc": 0.81826935, | |
| "epoch": 3.130081300813008, | |
| "grad_norm": 2.234375, | |
| "learning_rate": 3.4343794579172606e-05, | |
| "loss": 0.64635777, | |
| "memory(GiB)": 16.34, | |
| "step": 1155, | |
| "train_speed(iter/s)": 0.13924 | |
| }, | |
| { | |
| "acc": 0.79423141, | |
| "epoch": 3.1436314363143634, | |
| "grad_norm": 2.859375, | |
| "learning_rate": 3.380884450784593e-05, | |
| "loss": 0.71202464, | |
| "memory(GiB)": 16.34, | |
| "step": 1160, | |
| "train_speed(iter/s)": 0.139319 | |
| }, | |
| { | |
| "acc": 0.81948729, | |
| "epoch": 3.1571815718157183, | |
| "grad_norm": 2.046875, | |
| "learning_rate": 3.327389443651925e-05, | |
| "loss": 0.62771091, | |
| "memory(GiB)": 16.34, | |
| "step": 1165, | |
| "train_speed(iter/s)": 0.1394 | |
| }, | |
| { | |
| "acc": 0.85081501, | |
| "epoch": 3.1707317073170733, | |
| "grad_norm": 2.125, | |
| "learning_rate": 3.2738944365192576e-05, | |
| "loss": 0.55290155, | |
| "memory(GiB)": 16.34, | |
| "step": 1170, | |
| "train_speed(iter/s)": 0.139484 | |
| }, | |
| { | |
| "acc": 0.84410248, | |
| "epoch": 3.1842818428184283, | |
| "grad_norm": 2.140625, | |
| "learning_rate": 3.2203994293865906e-05, | |
| "loss": 0.57252893, | |
| "memory(GiB)": 16.34, | |
| "step": 1175, | |
| "train_speed(iter/s)": 0.139567 | |
| }, | |
| { | |
| "acc": 0.78565025, | |
| "epoch": 3.1978319783197833, | |
| "grad_norm": 3.203125, | |
| "learning_rate": 3.166904422253922e-05, | |
| "loss": 0.73501797, | |
| "memory(GiB)": 16.34, | |
| "step": 1180, | |
| "train_speed(iter/s)": 0.13965 | |
| }, | |
| { | |
| "acc": 0.78738356, | |
| "epoch": 3.2113821138211383, | |
| "grad_norm": 2.953125, | |
| "learning_rate": 3.113409415121255e-05, | |
| "loss": 0.72265806, | |
| "memory(GiB)": 16.34, | |
| "step": 1185, | |
| "train_speed(iter/s)": 0.139732 | |
| }, | |
| { | |
| "acc": 0.818221, | |
| "epoch": 3.2249322493224932, | |
| "grad_norm": 2.21875, | |
| "learning_rate": 3.0599144079885876e-05, | |
| "loss": 0.61062384, | |
| "memory(GiB)": 16.34, | |
| "step": 1190, | |
| "train_speed(iter/s)": 0.139813 | |
| }, | |
| { | |
| "acc": 0.82556448, | |
| "epoch": 3.238482384823848, | |
| "grad_norm": 2.3125, | |
| "learning_rate": 3.00641940085592e-05, | |
| "loss": 0.61080799, | |
| "memory(GiB)": 16.34, | |
| "step": 1195, | |
| "train_speed(iter/s)": 0.139893 | |
| }, | |
| { | |
| "acc": 0.82586126, | |
| "epoch": 3.252032520325203, | |
| "grad_norm": 2.34375, | |
| "learning_rate": 2.952924393723252e-05, | |
| "loss": 0.6164794, | |
| "memory(GiB)": 16.34, | |
| "step": 1200, | |
| "train_speed(iter/s)": 0.139974 | |
| }, | |
| { | |
| "epoch": 3.252032520325203, | |
| "eval_acc": 0.6044587062996115, | |
| "eval_loss": 1.8586076498031616, | |
| "eval_runtime": 44.1645, | |
| "eval_samples_per_second": 0.86, | |
| "eval_steps_per_second": 0.86, | |
| "step": 1200 | |
| }, | |
| { | |
| "acc": 0.82835131, | |
| "epoch": 3.265582655826558, | |
| "grad_norm": 6.46875, | |
| "learning_rate": 2.8994293865905846e-05, | |
| "loss": 0.61753302, | |
| "memory(GiB)": 16.34, | |
| "step": 1205, | |
| "train_speed(iter/s)": 0.139339 | |
| }, | |
| { | |
| "acc": 0.82603951, | |
| "epoch": 3.279132791327913, | |
| "grad_norm": 2.90625, | |
| "learning_rate": 2.8459343794579173e-05, | |
| "loss": 0.58360481, | |
| "memory(GiB)": 16.34, | |
| "step": 1210, | |
| "train_speed(iter/s)": 0.139423 | |
| }, | |
| { | |
| "acc": 0.79192553, | |
| "epoch": 3.292682926829268, | |
| "grad_norm": 2.5, | |
| "learning_rate": 2.7924393723252493e-05, | |
| "loss": 0.75090132, | |
| "memory(GiB)": 16.34, | |
| "step": 1215, | |
| "train_speed(iter/s)": 0.1395 | |
| }, | |
| { | |
| "acc": 0.83583679, | |
| "epoch": 3.306233062330623, | |
| "grad_norm": 2.15625, | |
| "learning_rate": 2.738944365192582e-05, | |
| "loss": 0.56217456, | |
| "memory(GiB)": 16.34, | |
| "step": 1220, | |
| "train_speed(iter/s)": 0.139581 | |
| }, | |
| { | |
| "acc": 0.80481911, | |
| "epoch": 3.3197831978319785, | |
| "grad_norm": 4.125, | |
| "learning_rate": 2.685449358059914e-05, | |
| "loss": 0.73073688, | |
| "memory(GiB)": 16.34, | |
| "step": 1225, | |
| "train_speed(iter/s)": 0.139659 | |
| }, | |
| { | |
| "acc": 0.8173975, | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 3.1875, | |
| "learning_rate": 2.6319543509272466e-05, | |
| "loss": 0.59854188, | |
| "memory(GiB)": 16.34, | |
| "step": 1230, | |
| "train_speed(iter/s)": 0.139735 | |
| }, | |
| { | |
| "acc": 0.88763266, | |
| "epoch": 3.3468834688346885, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 2.5784593437945793e-05, | |
| "loss": 0.3878314, | |
| "memory(GiB)": 16.34, | |
| "step": 1235, | |
| "train_speed(iter/s)": 0.139826 | |
| }, | |
| { | |
| "acc": 0.82199507, | |
| "epoch": 3.3604336043360434, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 2.5249643366619113e-05, | |
| "loss": 0.65350094, | |
| "memory(GiB)": 16.34, | |
| "step": 1240, | |
| "train_speed(iter/s)": 0.139901 | |
| }, | |
| { | |
| "acc": 0.83115234, | |
| "epoch": 3.3739837398373984, | |
| "grad_norm": 2.484375, | |
| "learning_rate": 2.471469329529244e-05, | |
| "loss": 0.61815515, | |
| "memory(GiB)": 16.34, | |
| "step": 1245, | |
| "train_speed(iter/s)": 0.13998 | |
| }, | |
| { | |
| "acc": 0.81266117, | |
| "epoch": 3.3875338753387534, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 2.417974322396576e-05, | |
| "loss": 0.65289016, | |
| "memory(GiB)": 16.34, | |
| "step": 1250, | |
| "train_speed(iter/s)": 0.140052 | |
| }, | |
| { | |
| "epoch": 3.3875338753387534, | |
| "eval_acc": 0.6037831447390644, | |
| "eval_loss": 1.893083095550537, | |
| "eval_runtime": 44.1639, | |
| "eval_samples_per_second": 0.86, | |
| "eval_steps_per_second": 0.86, | |
| "step": 1250 | |
| }, | |
| { | |
| "acc": 0.84776039, | |
| "epoch": 3.4010840108401084, | |
| "grad_norm": 2.046875, | |
| "learning_rate": 2.3644793152639086e-05, | |
| "loss": 0.53162961, | |
| "memory(GiB)": 16.34, | |
| "step": 1255, | |
| "train_speed(iter/s)": 0.139443 | |
| }, | |
| { | |
| "acc": 0.84318447, | |
| "epoch": 3.4146341463414633, | |
| "grad_norm": 2.234375, | |
| "learning_rate": 2.3109843081312406e-05, | |
| "loss": 0.52226324, | |
| "memory(GiB)": 16.34, | |
| "step": 1260, | |
| "train_speed(iter/s)": 0.139515 | |
| }, | |
| { | |
| "acc": 0.80521078, | |
| "epoch": 3.4281842818428183, | |
| "grad_norm": 2.25, | |
| "learning_rate": 2.2574893009985733e-05, | |
| "loss": 0.62338996, | |
| "memory(GiB)": 16.34, | |
| "step": 1265, | |
| "train_speed(iter/s)": 0.13959 | |
| }, | |
| { | |
| "acc": 0.83542995, | |
| "epoch": 3.4417344173441733, | |
| "grad_norm": 3.0625, | |
| "learning_rate": 2.203994293865906e-05, | |
| "loss": 0.60965805, | |
| "memory(GiB)": 16.34, | |
| "step": 1270, | |
| "train_speed(iter/s)": 0.139638 | |
| }, | |
| { | |
| "acc": 0.8333952, | |
| "epoch": 3.4552845528455283, | |
| "grad_norm": 2.625, | |
| "learning_rate": 2.150499286733238e-05, | |
| "loss": 0.60031776, | |
| "memory(GiB)": 16.34, | |
| "step": 1275, | |
| "train_speed(iter/s)": 0.139705 | |
| }, | |
| { | |
| "acc": 0.84367867, | |
| "epoch": 3.4688346883468837, | |
| "grad_norm": 1.859375, | |
| "learning_rate": 2.0970042796005706e-05, | |
| "loss": 0.54981174, | |
| "memory(GiB)": 16.34, | |
| "step": 1280, | |
| "train_speed(iter/s)": 0.13978 | |
| }, | |
| { | |
| "acc": 0.8121892, | |
| "epoch": 3.4823848238482387, | |
| "grad_norm": 2.046875, | |
| "learning_rate": 2.0435092724679026e-05, | |
| "loss": 0.67483497, | |
| "memory(GiB)": 16.34, | |
| "step": 1285, | |
| "train_speed(iter/s)": 0.139851 | |
| }, | |
| { | |
| "acc": 0.87883768, | |
| "epoch": 3.4959349593495936, | |
| "grad_norm": 1.6875, | |
| "learning_rate": 1.9900142653352353e-05, | |
| "loss": 0.45111437, | |
| "memory(GiB)": 16.34, | |
| "step": 1290, | |
| "train_speed(iter/s)": 0.139924 | |
| }, | |
| { | |
| "acc": 0.81952734, | |
| "epoch": 3.5094850948509486, | |
| "grad_norm": 2.640625, | |
| "learning_rate": 1.9365192582025676e-05, | |
| "loss": 0.66096244, | |
| "memory(GiB)": 16.34, | |
| "step": 1295, | |
| "train_speed(iter/s)": 0.139995 | |
| }, | |
| { | |
| "acc": 0.82709446, | |
| "epoch": 3.5230352303523036, | |
| "grad_norm": 2.140625, | |
| "learning_rate": 1.8830242510699e-05, | |
| "loss": 0.58278303, | |
| "memory(GiB)": 16.34, | |
| "step": 1300, | |
| "train_speed(iter/s)": 0.140068 | |
| }, | |
| { | |
| "epoch": 3.5230352303523036, | |
| "eval_acc": 0.603276473568654, | |
| "eval_loss": 1.8986655473709106, | |
| "eval_runtime": 44.3091, | |
| "eval_samples_per_second": 0.858, | |
| "eval_steps_per_second": 0.858, | |
| "step": 1300 | |
| }, | |
| { | |
| "acc": 0.83689537, | |
| "epoch": 3.5365853658536586, | |
| "grad_norm": 1.9453125, | |
| "learning_rate": 1.8295292439372323e-05, | |
| "loss": 0.55435977, | |
| "memory(GiB)": 16.34, | |
| "step": 1305, | |
| "train_speed(iter/s)": 0.139476 | |
| }, | |
| { | |
| "acc": 0.8525279, | |
| "epoch": 3.5501355013550135, | |
| "grad_norm": 2.21875, | |
| "learning_rate": 1.776034236804565e-05, | |
| "loss": 0.51831579, | |
| "memory(GiB)": 16.34, | |
| "step": 1310, | |
| "train_speed(iter/s)": 0.139549 | |
| }, | |
| { | |
| "acc": 0.83099527, | |
| "epoch": 3.5636856368563685, | |
| "grad_norm": 1.8828125, | |
| "learning_rate": 1.7225392296718973e-05, | |
| "loss": 0.61766572, | |
| "memory(GiB)": 16.34, | |
| "step": 1315, | |
| "train_speed(iter/s)": 0.139617 | |
| }, | |
| { | |
| "acc": 0.82115297, | |
| "epoch": 3.5772357723577235, | |
| "grad_norm": 2.765625, | |
| "learning_rate": 1.6690442225392296e-05, | |
| "loss": 0.62109137, | |
| "memory(GiB)": 16.34, | |
| "step": 1320, | |
| "train_speed(iter/s)": 0.13969 | |
| }, | |
| { | |
| "acc": 0.85132771, | |
| "epoch": 3.5907859078590785, | |
| "grad_norm": 2.421875, | |
| "learning_rate": 1.615549215406562e-05, | |
| "loss": 0.55242, | |
| "memory(GiB)": 16.34, | |
| "step": 1325, | |
| "train_speed(iter/s)": 0.139763 | |
| }, | |
| { | |
| "acc": 0.82085381, | |
| "epoch": 3.6043360433604335, | |
| "grad_norm": 2.8125, | |
| "learning_rate": 1.5620542082738943e-05, | |
| "loss": 0.62965093, | |
| "memory(GiB)": 16.34, | |
| "step": 1330, | |
| "train_speed(iter/s)": 0.139837 | |
| }, | |
| { | |
| "acc": 0.85592861, | |
| "epoch": 3.617886178861789, | |
| "grad_norm": 1.875, | |
| "learning_rate": 1.5085592011412266e-05, | |
| "loss": 0.54980545, | |
| "memory(GiB)": 16.34, | |
| "step": 1335, | |
| "train_speed(iter/s)": 0.139907 | |
| }, | |
| { | |
| "acc": 0.80786858, | |
| "epoch": 3.6314363143631434, | |
| "grad_norm": 2.328125, | |
| "learning_rate": 1.4550641940085591e-05, | |
| "loss": 0.67223382, | |
| "memory(GiB)": 16.34, | |
| "step": 1340, | |
| "train_speed(iter/s)": 0.139974 | |
| }, | |
| { | |
| "acc": 0.85045481, | |
| "epoch": 3.644986449864499, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 1.4015691868758914e-05, | |
| "loss": 0.50758219, | |
| "memory(GiB)": 16.34, | |
| "step": 1345, | |
| "train_speed(iter/s)": 0.140054 | |
| }, | |
| { | |
| "acc": 0.85536547, | |
| "epoch": 3.658536585365854, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 1.348074179743224e-05, | |
| "loss": 0.53356686, | |
| "memory(GiB)": 16.34, | |
| "step": 1350, | |
| "train_speed(iter/s)": 0.140124 | |
| }, | |
| { | |
| "epoch": 3.658536585365854, | |
| "eval_acc": 0.6010808984968755, | |
| "eval_loss": 1.8972593545913696, | |
| "eval_runtime": 44.2915, | |
| "eval_samples_per_second": 0.858, | |
| "eval_steps_per_second": 0.858, | |
| "step": 1350 | |
| }, | |
| { | |
| "acc": 0.8181016, | |
| "epoch": 3.6720867208672088, | |
| "grad_norm": 2.3125, | |
| "learning_rate": 1.2945791726105563e-05, | |
| "loss": 0.59544349, | |
| "memory(GiB)": 16.34, | |
| "step": 1355, | |
| "train_speed(iter/s)": 0.139555 | |
| }, | |
| { | |
| "acc": 0.87799835, | |
| "epoch": 3.6856368563685638, | |
| "grad_norm": 2.03125, | |
| "learning_rate": 1.2410841654778886e-05, | |
| "loss": 0.42857742, | |
| "memory(GiB)": 16.34, | |
| "step": 1360, | |
| "train_speed(iter/s)": 0.139631 | |
| }, | |
| { | |
| "acc": 0.81485729, | |
| "epoch": 3.6991869918699187, | |
| "grad_norm": 1.65625, | |
| "learning_rate": 1.187589158345221e-05, | |
| "loss": 0.64030995, | |
| "memory(GiB)": 16.34, | |
| "step": 1365, | |
| "train_speed(iter/s)": 0.139703 | |
| }, | |
| { | |
| "acc": 0.82011805, | |
| "epoch": 3.7127371273712737, | |
| "grad_norm": 2.125, | |
| "learning_rate": 1.1340941512125534e-05, | |
| "loss": 0.63428659, | |
| "memory(GiB)": 16.34, | |
| "step": 1370, | |
| "train_speed(iter/s)": 0.139771 | |
| }, | |
| { | |
| "acc": 0.78548884, | |
| "epoch": 3.7262872628726287, | |
| "grad_norm": 3.984375, | |
| "learning_rate": 1.0805991440798858e-05, | |
| "loss": 0.74558139, | |
| "memory(GiB)": 16.34, | |
| "step": 1375, | |
| "train_speed(iter/s)": 0.139836 | |
| }, | |
| { | |
| "acc": 0.85822716, | |
| "epoch": 3.7398373983739837, | |
| "grad_norm": 2.328125, | |
| "learning_rate": 1.0271041369472183e-05, | |
| "loss": 0.47486768, | |
| "memory(GiB)": 16.34, | |
| "step": 1380, | |
| "train_speed(iter/s)": 0.139902 | |
| }, | |
| { | |
| "acc": 0.83469105, | |
| "epoch": 3.7533875338753386, | |
| "grad_norm": 2.53125, | |
| "learning_rate": 9.736091298145506e-06, | |
| "loss": 0.56590233, | |
| "memory(GiB)": 16.34, | |
| "step": 1385, | |
| "train_speed(iter/s)": 0.139971 | |
| }, | |
| { | |
| "acc": 0.80486965, | |
| "epoch": 3.7669376693766936, | |
| "grad_norm": 3.28125, | |
| "learning_rate": 9.20114122681883e-06, | |
| "loss": 0.68021841, | |
| "memory(GiB)": 16.34, | |
| "step": 1390, | |
| "train_speed(iter/s)": 0.140036 | |
| }, | |
| { | |
| "acc": 0.84560328, | |
| "epoch": 3.7804878048780486, | |
| "grad_norm": 2.421875, | |
| "learning_rate": 8.666191155492154e-06, | |
| "loss": 0.52926989, | |
| "memory(GiB)": 16.34, | |
| "step": 1395, | |
| "train_speed(iter/s)": 0.140102 | |
| }, | |
| { | |
| "acc": 0.83154497, | |
| "epoch": 3.794037940379404, | |
| "grad_norm": 2.671875, | |
| "learning_rate": 8.131241084165478e-06, | |
| "loss": 0.5866107, | |
| "memory(GiB)": 16.34, | |
| "step": 1400, | |
| "train_speed(iter/s)": 0.14017 | |
| }, | |
| { | |
| "epoch": 3.794037940379404, | |
| "eval_acc": 0.6036142543489276, | |
| "eval_loss": 1.8983793258666992, | |
| "eval_runtime": 44.4268, | |
| "eval_samples_per_second": 0.855, | |
| "eval_steps_per_second": 0.855, | |
| "step": 1400 | |
| }, | |
| { | |
| "acc": 0.84980259, | |
| "epoch": 3.8075880758807585, | |
| "grad_norm": 2.859375, | |
| "learning_rate": 7.596291012838801e-06, | |
| "loss": 0.56505013, | |
| "memory(GiB)": 16.34, | |
| "step": 1405, | |
| "train_speed(iter/s)": 0.139621 | |
| }, | |
| { | |
| "acc": 0.80126667, | |
| "epoch": 3.821138211382114, | |
| "grad_norm": 2.21875, | |
| "learning_rate": 7.061340941512125e-06, | |
| "loss": 0.67650762, | |
| "memory(GiB)": 16.34, | |
| "step": 1410, | |
| "train_speed(iter/s)": 0.139688 | |
| }, | |
| { | |
| "acc": 0.79974804, | |
| "epoch": 3.834688346883469, | |
| "grad_norm": 2.3125, | |
| "learning_rate": 6.5263908701854486e-06, | |
| "loss": 0.743855, | |
| "memory(GiB)": 16.34, | |
| "step": 1415, | |
| "train_speed(iter/s)": 0.139749 | |
| }, | |
| { | |
| "acc": 0.84023552, | |
| "epoch": 3.848238482384824, | |
| "grad_norm": 2.234375, | |
| "learning_rate": 5.991440798858773e-06, | |
| "loss": 0.57722979, | |
| "memory(GiB)": 16.34, | |
| "step": 1420, | |
| "train_speed(iter/s)": 0.139813 | |
| }, | |
| { | |
| "acc": 0.82472792, | |
| "epoch": 3.861788617886179, | |
| "grad_norm": 2.484375, | |
| "learning_rate": 5.456490727532097e-06, | |
| "loss": 0.56691737, | |
| "memory(GiB)": 16.34, | |
| "step": 1425, | |
| "train_speed(iter/s)": 0.139877 | |
| }, | |
| { | |
| "acc": 0.83573132, | |
| "epoch": 3.875338753387534, | |
| "grad_norm": 1.9765625, | |
| "learning_rate": 4.92154065620542e-06, | |
| "loss": 0.60298834, | |
| "memory(GiB)": 16.34, | |
| "step": 1430, | |
| "train_speed(iter/s)": 0.13994 | |
| }, | |
| { | |
| "acc": 0.83143454, | |
| "epoch": 3.888888888888889, | |
| "grad_norm": 1.9453125, | |
| "learning_rate": 4.386590584878744e-06, | |
| "loss": 0.61832671, | |
| "memory(GiB)": 16.34, | |
| "step": 1435, | |
| "train_speed(iter/s)": 0.140003 | |
| }, | |
| { | |
| "acc": 0.83209105, | |
| "epoch": 3.902439024390244, | |
| "grad_norm": 3.265625, | |
| "learning_rate": 3.851640513552068e-06, | |
| "loss": 0.6038147, | |
| "memory(GiB)": 16.34, | |
| "step": 1440, | |
| "train_speed(iter/s)": 0.140065 | |
| }, | |
| { | |
| "acc": 0.79715924, | |
| "epoch": 3.915989159891599, | |
| "grad_norm": 2.015625, | |
| "learning_rate": 3.316690442225392e-06, | |
| "loss": 0.70867634, | |
| "memory(GiB)": 16.34, | |
| "step": 1445, | |
| "train_speed(iter/s)": 0.140125 | |
| }, | |
| { | |
| "acc": 0.85556803, | |
| "epoch": 3.9295392953929538, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 2.781740370898716e-06, | |
| "loss": 0.52723417, | |
| "memory(GiB)": 16.34, | |
| "step": 1450, | |
| "train_speed(iter/s)": 0.140182 | |
| }, | |
| { | |
| "epoch": 3.9295392953929538, | |
| "eval_acc": 0.6029386927883803, | |
| "eval_loss": 1.8965932130813599, | |
| "eval_runtime": 44.4125, | |
| "eval_samples_per_second": 0.856, | |
| "eval_steps_per_second": 0.856, | |
| "step": 1450 | |
| }, | |
| { | |
| "acc": 0.83306837, | |
| "epoch": 3.943089430894309, | |
| "grad_norm": 2.25, | |
| "learning_rate": 2.2467902995720398e-06, | |
| "loss": 0.55906692, | |
| "memory(GiB)": 16.34, | |
| "step": 1455, | |
| "train_speed(iter/s)": 0.139645 | |
| }, | |
| { | |
| "acc": 0.81574478, | |
| "epoch": 3.9566395663956637, | |
| "grad_norm": 2.6875, | |
| "learning_rate": 1.7118402282453637e-06, | |
| "loss": 0.62746248, | |
| "memory(GiB)": 16.34, | |
| "step": 1460, | |
| "train_speed(iter/s)": 0.139708 | |
| }, | |
| { | |
| "acc": 0.83905201, | |
| "epoch": 3.970189701897019, | |
| "grad_norm": 2.46875, | |
| "learning_rate": 1.1768901569186875e-06, | |
| "loss": 0.56978045, | |
| "memory(GiB)": 16.34, | |
| "step": 1465, | |
| "train_speed(iter/s)": 0.139771 | |
| }, | |
| { | |
| "acc": 0.85569115, | |
| "epoch": 3.983739837398374, | |
| "grad_norm": 1.90625, | |
| "learning_rate": 6.419400855920114e-07, | |
| "loss": 0.48882861, | |
| "memory(GiB)": 16.34, | |
| "step": 1470, | |
| "train_speed(iter/s)": 0.139834 | |
| }, | |
| { | |
| "acc": 0.82850809, | |
| "epoch": 3.997289972899729, | |
| "grad_norm": 1.828125, | |
| "learning_rate": 1.0699001426533523e-07, | |
| "loss": 0.56869435, | |
| "memory(GiB)": 16.34, | |
| "step": 1475, | |
| "train_speed(iter/s)": 0.139901 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_acc": 0.6037831447390644, | |
| "eval_loss": 1.8978389501571655, | |
| "eval_runtime": 44.2826, | |
| "eval_samples_per_second": 0.858, | |
| "eval_steps_per_second": 0.858, | |
| "step": 1476 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 1476, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 369, | |
| "total_flos": 1.991427194376192e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |