| { | |
| "best_metric": 3.165127754211426, | |
| "best_model_checkpoint": "/data1/attanasiog/babylm/roberta-tiny-2l-10M/checkpoint-4400", | |
| "epoch": 94.78098908156711, | |
| "global_step": 4550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8e-05, | |
| "loss": 10.5161, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00016, | |
| "loss": 9.1097, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00024, | |
| "loss": 7.8514, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00032, | |
| "loss": 7.3238, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.0004, | |
| "loss": 7.7619, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "eval_accuracy": 0.07476398255519703, | |
| "eval_loss": 7.233829975128174, | |
| "eval_runtime": 127.6677, | |
| "eval_samples_per_second": 188.419, | |
| "eval_steps_per_second": 5.89, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.000399995625676045, | |
| "loss": 7.1368, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.0003999825028955268, | |
| "loss": 6.9374, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.0003999606322324786, | |
| "loss": 6.8035, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 0.0003999300146435939, | |
| "loss": 6.6903, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.00039989065146818525, | |
| "loss": 7.0524, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_accuracy": 0.1331419040979615, | |
| "eval_loss": 6.625187873840332, | |
| "eval_runtime": 127.7949, | |
| "eval_samples_per_second": 188.231, | |
| "eval_steps_per_second": 5.884, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 0.0003998425444281255, | |
| "loss": 6.5528, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00039978569562777234, | |
| "loss": 6.5145, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0003997201075538765, | |
| "loss": 6.4642, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.0003996457830754729, | |
| "loss": 6.4561, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.00039956272544375493, | |
| "loss": 6.8423, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 0.14625706249076864, | |
| "eval_loss": 6.462179183959961, | |
| "eval_runtime": 127.8455, | |
| "eval_samples_per_second": 188.157, | |
| "eval_steps_per_second": 5.882, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.00039947093829193245, | |
| "loss": 6.3841, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 0.00039937042563507283, | |
| "loss": 6.3875, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 0.00039926119186992537, | |
| "loss": 6.3843, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 0.0003991432417747288, | |
| "loss": 6.3505, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 0.0003990165805090023, | |
| "loss": 6.7298, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "eval_accuracy": 0.1487513607434242, | |
| "eval_loss": 6.397099018096924, | |
| "eval_runtime": 127.9548, | |
| "eval_samples_per_second": 187.996, | |
| "eval_steps_per_second": 5.877, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 0.00039888121361332003, | |
| "loss": 6.3075, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 0.0003987371470090686, | |
| "loss": 6.3248, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 0.00039858438699818784, | |
| "loss": 6.2949, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 0.0003984229402628956, | |
| "loss": 6.2768, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 0.00039825281386539503, | |
| "loss": 6.669, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "eval_accuracy": 0.15192705186935002, | |
| "eval_loss": 6.362815856933594, | |
| "eval_runtime": 127.8986, | |
| "eval_samples_per_second": 188.079, | |
| "eval_steps_per_second": 5.88, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 0.000398074015247566, | |
| "loss": 6.257, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 0.0003978865522306392, | |
| "loss": 6.2485, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "learning_rate": 0.0003976904330148543, | |
| "loss": 6.252, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 0.00039748566617910113, | |
| "loss": 6.6549, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 0.0003972722606805445, | |
| "loss": 6.2038, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "eval_accuracy": 0.15184154910887893, | |
| "eval_loss": 6.337147235870361, | |
| "eval_runtime": 128.0686, | |
| "eval_samples_per_second": 187.829, | |
| "eval_steps_per_second": 5.872, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.00039705022585423216, | |
| "loss": 6.2262, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 6.66, | |
| "learning_rate": 0.0003968195714126868, | |
| "loss": 6.2023, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 0.00039658030744548075, | |
| "loss": 6.2053, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 0.0003963324444187952, | |
| "loss": 6.6064, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "learning_rate": 0.0003960759931749619, | |
| "loss": 6.1783, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "eval_accuracy": 0.15316276995162847, | |
| "eval_loss": 6.311531066894531, | |
| "eval_runtime": 127.9978, | |
| "eval_samples_per_second": 187.933, | |
| "eval_steps_per_second": 5.875, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 0.00039581096493198893, | |
| "loss": 6.178, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "learning_rate": 0.0003955373712830703, | |
| "loss": 6.1784, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 7.9, | |
| "learning_rate": 0.00039525522419607854, | |
| "loss": 6.1739, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 0.0003949645360130412, | |
| "loss": 6.5644, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 0.0003946653194496012, | |
| "loss": 6.1459, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "eval_accuracy": 0.15298167344807118, | |
| "eval_loss": 6.292238712310791, | |
| "eval_runtime": 218.0011, | |
| "eval_samples_per_second": 110.343, | |
| "eval_steps_per_second": 3.45, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "learning_rate": 0.00039435758759446025, | |
| "loss": 6.1514, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "learning_rate": 0.00039404135390880664, | |
| "loss": 6.1335, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 0.0003937166322257262, | |
| "loss": 6.1613, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 0.00039338343674959745, | |
| "loss": 6.5555, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 9.37, | |
| "learning_rate": 0.00039304178205546976, | |
| "loss": 6.1096, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 9.37, | |
| "eval_accuracy": 0.15364433508531855, | |
| "eval_loss": 6.269557952880859, | |
| "eval_runtime": 245.4855, | |
| "eval_samples_per_second": 97.989, | |
| "eval_steps_per_second": 3.063, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 0.00039269168308842634, | |
| "loss": 6.1131, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 0.00039233315516293006, | |
| "loss": 6.1172, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "learning_rate": 0.00039196621396215403, | |
| "loss": 6.0984, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 10.21, | |
| "learning_rate": 0.000391590875537295, | |
| "loss": 6.494, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 0.00039120715630687155, | |
| "loss": 6.0745, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "eval_accuracy": 0.15413396308915142, | |
| "eval_loss": 6.25447416305542, | |
| "eval_runtime": 243.2903, | |
| "eval_samples_per_second": 98.874, | |
| "eval_steps_per_second": 3.091, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 0.000390815073056006, | |
| "loss": 6.0953, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 10.82, | |
| "learning_rate": 0.00039041464293568983, | |
| "loss": 6.0869, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 0.00039000588346203374, | |
| "loss": 6.4846, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 0.0003895888125155014, | |
| "loss": 6.0673, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 0.00038916344834012695, | |
| "loss": 6.0689, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "eval_accuracy": 0.15334541266013718, | |
| "eval_loss": 6.24962854385376, | |
| "eval_runtime": 238.4372, | |
| "eval_samples_per_second": 100.886, | |
| "eval_steps_per_second": 3.154, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 11.66, | |
| "learning_rate": 0.00038872980954271757, | |
| "loss": 6.0805, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 0.00038828791509203895, | |
| "loss": 6.0632, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 12.08, | |
| "learning_rate": 0.00038783778431798597, | |
| "loss": 6.4656, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 12.29, | |
| "learning_rate": 0.0003873794369107369, | |
| "loss": 6.0445, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "learning_rate": 0.0003869128929198922, | |
| "loss": 6.0562, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "eval_accuracy": 0.15423183376396205, | |
| "eval_loss": 6.231264114379883, | |
| "eval_runtime": 244.3792, | |
| "eval_samples_per_second": 98.433, | |
| "eval_steps_per_second": 3.077, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 12.7, | |
| "learning_rate": 0.0003864381727535973, | |
| "loss": 6.0553, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 0.00038595529717765027, | |
| "loss": 6.0595, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 0.0003854642873145931, | |
| "loss": 6.445, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 0.00038496516464278776, | |
| "loss": 6.0285, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 13.53, | |
| "learning_rate": 0.00038445795099547697, | |
| "loss": 6.0324, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 13.53, | |
| "eval_accuracy": 0.15358873091243086, | |
| "eval_loss": 6.224751949310303, | |
| "eval_runtime": 128.0464, | |
| "eval_samples_per_second": 187.862, | |
| "eval_steps_per_second": 5.873, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 13.74, | |
| "learning_rate": 0.0003839426685598287, | |
| "loss": 6.0219, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "learning_rate": 0.000383419339875966, | |
| "loss": 6.0379, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 14.16, | |
| "learning_rate": 0.00038288798783598087, | |
| "loss": 6.416, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 14.37, | |
| "learning_rate": 0.0003823486356829329, | |
| "loss": 5.9984, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 14.58, | |
| "learning_rate": 0.0003818013070098325, | |
| "loss": 5.9907, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 14.58, | |
| "eval_accuracy": 0.15438304013536042, | |
| "eval_loss": 6.217936038970947, | |
| "eval_runtime": 128.0604, | |
| "eval_samples_per_second": 187.841, | |
| "eval_steps_per_second": 5.872, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "learning_rate": 0.0003812460257586089, | |
| "loss": 6.0038, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "learning_rate": 0.000380682816219063, | |
| "loss": 6.0321, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 15.21, | |
| "learning_rate": 0.00038011170302780446, | |
| "loss": 6.3685, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 15.41, | |
| "learning_rate": 0.00037953271116717444, | |
| "loss": 5.9825, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "learning_rate": 0.0003789458659641527, | |
| "loss": 5.9683, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "eval_accuracy": 0.1545422454380471, | |
| "eval_loss": 6.183169364929199, | |
| "eval_runtime": 128.0339, | |
| "eval_samples_per_second": 187.88, | |
| "eval_steps_per_second": 5.873, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 15.82, | |
| "learning_rate": 0.0003783511930892495, | |
| "loss": 5.9712, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 0.00037774871855538275, | |
| "loss": 6.355, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "learning_rate": 0.00037713846871674045, | |
| "loss": 5.9361, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 16.45, | |
| "learning_rate": 0.0003765204702676274, | |
| "loss": 5.9281, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 16.66, | |
| "learning_rate": 0.0003758947502412978, | |
| "loss": 5.9236, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 16.66, | |
| "eval_accuracy": 0.15502017774172816, | |
| "eval_loss": 6.141255855560303, | |
| "eval_runtime": 128.0268, | |
| "eval_samples_per_second": 187.89, | |
| "eval_steps_per_second": 5.874, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 16.86, | |
| "learning_rate": 0.0003752613360087727, | |
| "loss": 5.9288, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 17.08, | |
| "learning_rate": 0.00037462025527764265, | |
| "loss": 6.2842, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 17.29, | |
| "learning_rate": 0.00037397153609085553, | |
| "loss": 5.8852, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 17.49, | |
| "learning_rate": 0.0003733152068254901, | |
| "loss": 5.8779, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 17.7, | |
| "learning_rate": 0.00037265129619151483, | |
| "loss": 5.8808, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 17.7, | |
| "eval_accuracy": 0.15577406052421716, | |
| "eval_loss": 6.089950084686279, | |
| "eval_runtime": 128.2202, | |
| "eval_samples_per_second": 187.607, | |
| "eval_steps_per_second": 5.865, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 17.9, | |
| "learning_rate": 0.00037197983323053143, | |
| "loss": 5.871, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "learning_rate": 0.00037130084731450515, | |
| "loss": 6.2293, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 18.33, | |
| "learning_rate": 0.0003706143681444795, | |
| "loss": 5.8282, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 18.53, | |
| "learning_rate": 0.0003699204257492774, | |
| "loss": 5.85, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 18.74, | |
| "learning_rate": 0.0003692190504841871, | |
| "loss": 5.8392, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 18.74, | |
| "eval_accuracy": 0.15657369591332176, | |
| "eval_loss": 6.054327487945557, | |
| "eval_runtime": 128.0437, | |
| "eval_samples_per_second": 187.866, | |
| "eval_steps_per_second": 5.873, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 18.95, | |
| "learning_rate": 0.00036851027302963493, | |
| "loss": 5.8393, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 19.16, | |
| "learning_rate": 0.00036779412438984294, | |
| "loss": 6.1961, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 19.37, | |
| "learning_rate": 0.0003670706358914725, | |
| "loss": 5.8161, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "learning_rate": 0.0003663398391822543, | |
| "loss": 5.7886, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 19.78, | |
| "learning_rate": 0.00036560176622960403, | |
| "loss": 5.7962, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 19.78, | |
| "eval_accuracy": 0.15750512966626293, | |
| "eval_loss": 6.022204399108887, | |
| "eval_runtime": 127.9436, | |
| "eval_samples_per_second": 188.012, | |
| "eval_steps_per_second": 5.878, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "learning_rate": 0.00036485644931922353, | |
| "loss": 5.7823, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 20.21, | |
| "learning_rate": 0.0003641039210536889, | |
| "loss": 6.1533, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 20.41, | |
| "learning_rate": 0.0003633442143510245, | |
| "loss": 5.7526, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 20.62, | |
| "learning_rate": 0.00036257736244326246, | |
| "loss": 5.7454, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 20.82, | |
| "learning_rate": 0.0003618033988749895, | |
| "loss": 5.7473, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 20.82, | |
| "eval_accuracy": 0.16172566383651218, | |
| "eval_loss": 5.947088718414307, | |
| "eval_runtime": 128.1904, | |
| "eval_samples_per_second": 187.65, | |
| "eval_steps_per_second": 5.866, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 21.04, | |
| "learning_rate": 0.0003610223575018795, | |
| "loss": 6.0948, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "learning_rate": 0.00036023427248921215, | |
| "loss": 5.6776, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 21.45, | |
| "learning_rate": 0.0003594391783103792, | |
| "loss": 5.6479, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 21.66, | |
| "learning_rate": 0.00035863710974537563, | |
| "loss": 5.6245, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 21.86, | |
| "learning_rate": 0.00035782810187927875, | |
| "loss": 5.5787, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 21.86, | |
| "eval_accuracy": 0.18910485199927482, | |
| "eval_loss": 5.7037835121154785, | |
| "eval_runtime": 128.2758, | |
| "eval_samples_per_second": 187.526, | |
| "eval_steps_per_second": 5.862, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 22.08, | |
| "learning_rate": 0.0003570121901007136, | |
| "loss": 5.8678, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 22.29, | |
| "learning_rate": 0.0003561894101003044, | |
| "loss": 5.4587, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 22.49, | |
| "learning_rate": 0.00035535979786911396, | |
| "loss": 5.3982, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 22.7, | |
| "learning_rate": 0.00035452338969706876, | |
| "loss": 5.305, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 22.9, | |
| "learning_rate": 0.00035368022217137184, | |
| "loss": 5.2316, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 22.9, | |
| "eval_accuracy": 0.23819718901662149, | |
| "eval_loss": 5.270751476287842, | |
| "eval_runtime": 128.1695, | |
| "eval_samples_per_second": 187.681, | |
| "eval_steps_per_second": 5.867, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 23.12, | |
| "learning_rate": 0.00035283033217490227, | |
| "loss": 5.4202, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 23.33, | |
| "learning_rate": 0.00035197375688460176, | |
| "loss": 4.9911, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 23.53, | |
| "learning_rate": 0.0003511105337698484, | |
| "loss": 4.8741, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 23.74, | |
| "learning_rate": 0.0003502407005908177, | |
| "loss": 4.7582, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 23.95, | |
| "learning_rate": 0.0003493642953968308, | |
| "loss": 4.6613, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 23.95, | |
| "eval_accuracy": 0.29748286605712254, | |
| "eval_loss": 4.707459926605225, | |
| "eval_runtime": 131.3634, | |
| "eval_samples_per_second": 183.118, | |
| "eval_steps_per_second": 5.725, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 24.16, | |
| "learning_rate": 0.00034848135652469, | |
| "loss": 4.8536, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 24.37, | |
| "learning_rate": 0.00034759192259700196, | |
| "loss": 4.4822, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 24.58, | |
| "learning_rate": 0.000346696032520488, | |
| "loss": 4.4126, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 24.78, | |
| "learning_rate": 0.00034579372548428235, | |
| "loss": 4.3707, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "learning_rate": 0.00034488504095821784, | |
| "loss": 4.3006, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "eval_accuracy": 0.3221731878424314, | |
| "eval_loss": 4.417978763580322, | |
| "eval_runtime": 144.7447, | |
| "eval_samples_per_second": 166.189, | |
| "eval_steps_per_second": 5.195, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 25.21, | |
| "learning_rate": 0.0003439700186910993, | |
| "loss": 4.5185, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 25.41, | |
| "learning_rate": 0.00034304869870896513, | |
| "loss": 4.2011, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 25.62, | |
| "learning_rate": 0.00034212112131333587, | |
| "loss": 4.1513, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 25.82, | |
| "learning_rate": 0.0003411873270794518, | |
| "loss": 4.1584, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 26.04, | |
| "learning_rate": 0.00034024735685449773, | |
| "loss": 4.3754, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 26.04, | |
| "eval_accuracy": 0.33853883405739793, | |
| "eval_loss": 4.238345146179199, | |
| "eval_runtime": 144.4875, | |
| "eval_samples_per_second": 166.485, | |
| "eval_steps_per_second": 5.205, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 26.25, | |
| "learning_rate": 0.00033930125175581647, | |
| "loss": 4.0477, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 26.45, | |
| "learning_rate": 0.0003383490531691099, | |
| "loss": 4.0339, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 26.66, | |
| "learning_rate": 0.0003373908027466289, | |
| "loss": 4.0184, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 26.86, | |
| "learning_rate": 0.00033642654240535134, | |
| "loss": 3.9835, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 27.08, | |
| "learning_rate": 0.00033545631432514825, | |
| "loss": 4.2531, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 27.08, | |
| "eval_accuracy": 0.34910193403738843, | |
| "eval_loss": 4.1157379150390625, | |
| "eval_runtime": 144.9052, | |
| "eval_samples_per_second": 166.005, | |
| "eval_steps_per_second": 5.19, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 27.29, | |
| "learning_rate": 0.00033448016094693895, | |
| "loss": 3.9085, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 27.49, | |
| "learning_rate": 0.0003334981249708345, | |
| "loss": 3.9205, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 27.7, | |
| "learning_rate": 0.00033251024935427, | |
| "loss": 3.8786, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 27.9, | |
| "learning_rate": 0.0003315165773101249, | |
| "loss": 3.8839, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 28.12, | |
| "learning_rate": 0.00033051715230483374, | |
| "loss": 4.0987, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 28.12, | |
| "eval_accuracy": 0.3577664889956034, | |
| "eval_loss": 4.0197062492370605, | |
| "eval_runtime": 128.1918, | |
| "eval_samples_per_second": 187.649, | |
| "eval_steps_per_second": 5.866, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 28.33, | |
| "learning_rate": 0.0003295120180564838, | |
| "loss": 3.8043, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 28.53, | |
| "learning_rate": 0.00032850121853290334, | |
| "loss": 3.8048, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 28.74, | |
| "learning_rate": 0.000327484797949738, | |
| "loss": 3.8127, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 28.95, | |
| "learning_rate": 0.00032646280076851684, | |
| "loss": 3.7773, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 29.16, | |
| "learning_rate": 0.0003254352716947074, | |
| "loss": 4.0045, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 29.16, | |
| "eval_accuracy": 0.3656404849021274, | |
| "eval_loss": 3.9503591060638428, | |
| "eval_runtime": 128.0514, | |
| "eval_samples_per_second": 187.854, | |
| "eval_steps_per_second": 5.873, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 29.37, | |
| "learning_rate": 0.0003244022556757602, | |
| "loss": 3.7454, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 29.58, | |
| "learning_rate": 0.0003233637978991422, | |
| "loss": 3.7151, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 29.78, | |
| "learning_rate": 0.00032231994379036086, | |
| "loss": 3.7243, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 29.99, | |
| "learning_rate": 0.0003212707390109765, | |
| "loss": 3.691, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 30.21, | |
| "learning_rate": 0.00032021622945660504, | |
| "loss": 3.9145, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 30.21, | |
| "eval_accuracy": 0.37178706581714294, | |
| "eval_loss": 3.8818981647491455, | |
| "eval_runtime": 128.0667, | |
| "eval_samples_per_second": 187.832, | |
| "eval_steps_per_second": 5.872, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 30.41, | |
| "learning_rate": 0.0003191564612549106, | |
| "loss": 3.6485, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 30.62, | |
| "learning_rate": 0.0003180914807635874, | |
| "loss": 3.6517, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 30.82, | |
| "learning_rate": 0.00031702133456833236, | |
| "loss": 3.6453, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 31.04, | |
| "learning_rate": 0.00031594606948080663, | |
| "loss": 3.8657, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 31.25, | |
| "learning_rate": 0.00031486573253658874, | |
| "loss": 3.5808, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 31.25, | |
| "eval_accuracy": 0.3780561563488311, | |
| "eval_loss": 3.827902317047119, | |
| "eval_runtime": 128.1177, | |
| "eval_samples_per_second": 187.757, | |
| "eval_steps_per_second": 5.87, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 31.45, | |
| "learning_rate": 0.00031378037099311627, | |
| "loss": 3.5983, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 31.66, | |
| "learning_rate": 0.00031269003232761933, | |
| "loss": 3.5947, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 31.86, | |
| "learning_rate": 0.0003115947642350433, | |
| "loss": 3.5773, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 32.08, | |
| "learning_rate": 0.00031049461462596267, | |
| "loss": 3.7941, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 32.29, | |
| "learning_rate": 0.00030938963162448544, | |
| "loss": 3.5354, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 32.29, | |
| "eval_accuracy": 0.3825753450782098, | |
| "eval_loss": 3.7829582691192627, | |
| "eval_runtime": 128.1881, | |
| "eval_samples_per_second": 187.654, | |
| "eval_steps_per_second": 5.866, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 32.49, | |
| "learning_rate": 0.0003082798635661476, | |
| "loss": 3.5355, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 32.7, | |
| "learning_rate": 0.0003071653589957993, | |
| "loss": 3.5166, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "learning_rate": 0.000306046166665481, | |
| "loss": 3.5295, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 33.12, | |
| "learning_rate": 0.00030492233553229076, | |
| "loss": 3.7281, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 0.00030379391475624304, | |
| "loss": 3.4788, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "eval_accuracy": 0.3872195585541312, | |
| "eval_loss": 3.7400357723236084, | |
| "eval_runtime": 128.395, | |
| "eval_samples_per_second": 187.352, | |
| "eval_steps_per_second": 5.857, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 33.53, | |
| "learning_rate": 0.0003026609536981183, | |
| "loss": 3.4722, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 33.74, | |
| "learning_rate": 0.0003015235019173034, | |
| "loss": 3.4888, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 33.95, | |
| "learning_rate": 0.00030038160916962404, | |
| "loss": 3.472, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 34.16, | |
| "learning_rate": 0.00029923532540516843, | |
| "loss": 3.6802, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 34.37, | |
| "learning_rate": 0.00029808470076610167, | |
| "loss": 3.4315, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 34.37, | |
| "eval_accuracy": 0.39113526163842305, | |
| "eval_loss": 3.702760696411133, | |
| "eval_runtime": 147.8935, | |
| "eval_samples_per_second": 162.651, | |
| "eval_steps_per_second": 5.085, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 34.58, | |
| "learning_rate": 0.00029692978558447305, | |
| "loss": 3.4258, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 34.78, | |
| "learning_rate": 0.0002957706303800139, | |
| "loss": 3.4325, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 34.99, | |
| "learning_rate": 0.0002946072858579282, | |
| "loss": 3.418, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 35.21, | |
| "learning_rate": 0.0002934398029066739, | |
| "loss": 3.6162, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 35.41, | |
| "learning_rate": 0.0002922682325957376, | |
| "loss": 3.3906, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 35.41, | |
| "eval_accuracy": 0.3955544302044244, | |
| "eval_loss": 3.6628527641296387, | |
| "eval_runtime": 147.4398, | |
| "eval_samples_per_second": 163.151, | |
| "eval_steps_per_second": 5.1, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 35.62, | |
| "learning_rate": 0.00029109262617339987, | |
| "loss": 3.3731, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 35.82, | |
| "learning_rate": 0.0002899130350644941, | |
| "loss": 3.381, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 36.04, | |
| "learning_rate": 0.00028872951086815685, | |
| "loss": 3.608, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 36.25, | |
| "learning_rate": 0.00028754210535557036, | |
| "loss": 3.3345, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 36.45, | |
| "learning_rate": 0.00028635087046769857, | |
| "loss": 3.3508, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 36.45, | |
| "eval_accuracy": 0.3984451691466028, | |
| "eval_loss": 3.6344361305236816, | |
| "eval_runtime": 147.716, | |
| "eval_samples_per_second": 162.846, | |
| "eval_steps_per_second": 5.091, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 36.66, | |
| "learning_rate": 0.00028515585831301456, | |
| "loss": 3.3503, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 36.86, | |
| "learning_rate": 0.0002839571211652212, | |
| "loss": 3.3494, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 37.08, | |
| "learning_rate": 0.00028275471146096466, | |
| "loss": 3.539, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 37.29, | |
| "learning_rate": 0.00028154868179754074, | |
| "loss": 3.3145, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 37.49, | |
| "learning_rate": 0.0002803390849305939, | |
| "loss": 3.288, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 37.49, | |
| "eval_accuracy": 0.4019129197606658, | |
| "eval_loss": 3.6045737266540527, | |
| "eval_runtime": 128.0819, | |
| "eval_samples_per_second": 187.81, | |
| "eval_steps_per_second": 5.871, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 37.7, | |
| "learning_rate": 0.0002791259737718097, | |
| "loss": 3.318, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 37.9, | |
| "learning_rate": 0.0002779094013866001, | |
| "loss": 3.3005, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 38.12, | |
| "learning_rate": 0.00027668942099178234, | |
| "loss": 3.4959, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 38.33, | |
| "learning_rate": 0.00027546608595325117, | |
| "loss": 3.2771, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 38.53, | |
| "learning_rate": 0.00027423944978364416, | |
| "loss": 3.2678, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 38.53, | |
| "eval_accuracy": 0.40528409278500616, | |
| "eval_loss": 3.5798938274383545, | |
| "eval_runtime": 128.148, | |
| "eval_samples_per_second": 187.713, | |
| "eval_steps_per_second": 5.868, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 38.74, | |
| "learning_rate": 0.00027300956614000115, | |
| "loss": 3.2707, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 38.95, | |
| "learning_rate": 0.00027177648882141704, | |
| "loss": 3.276, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 39.16, | |
| "learning_rate": 0.0002705402717666883, | |
| "loss": 3.4633, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 39.37, | |
| "learning_rate": 0.00026930096905195363, | |
| "loss": 3.2392, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 39.58, | |
| "learning_rate": 0.00026805863488832865, | |
| "loss": 3.2382, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 39.58, | |
| "eval_accuracy": 0.40741369917263087, | |
| "eval_loss": 3.5548534393310547, | |
| "eval_runtime": 128.15, | |
| "eval_samples_per_second": 187.71, | |
| "eval_steps_per_second": 5.868, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 39.78, | |
| "learning_rate": 0.00026681332361953424, | |
| "loss": 3.2304, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 39.99, | |
| "learning_rate": 0.0002655650897195195, | |
| "loss": 3.24, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 40.21, | |
| "learning_rate": 0.0002643139877900791, | |
| "loss": 3.4143, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 40.41, | |
| "learning_rate": 0.00026306007255846436, | |
| "loss": 3.203, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 40.62, | |
| "learning_rate": 0.00026180339887498953, | |
| "loss": 3.2151, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 40.62, | |
| "eval_accuracy": 0.41034103588846577, | |
| "eval_loss": 3.5284957885742188, | |
| "eval_runtime": 128.1661, | |
| "eval_samples_per_second": 187.686, | |
| "eval_steps_per_second": 5.867, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 40.82, | |
| "learning_rate": 0.00026054402171063267, | |
| "loss": 3.2063, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 41.04, | |
| "learning_rate": 0.0002592819961546308, | |
| "loss": 3.4173, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 41.25, | |
| "learning_rate": 0.00025801737741207005, | |
| "loss": 3.1796, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 41.45, | |
| "learning_rate": 0.000256750220801471, | |
| "loss": 3.1799, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 41.66, | |
| "learning_rate": 0.0002554805817523689, | |
| "loss": 3.1777, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 41.66, | |
| "eval_accuracy": 0.41320715004942615, | |
| "eval_loss": 3.506920337677002, | |
| "eval_runtime": 128.1475, | |
| "eval_samples_per_second": 187.713, | |
| "eval_steps_per_second": 5.868, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 41.86, | |
| "learning_rate": 0.0002542085158028889, | |
| "loss": 3.1791, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 42.08, | |
| "learning_rate": 0.00025293407859731633, | |
| "loss": 3.363, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 42.29, | |
| "learning_rate": 0.00025165732588366334, | |
| "loss": 3.1381, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 42.49, | |
| "learning_rate": 0.00025037831351122967, | |
| "loss": 3.1556, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 42.7, | |
| "learning_rate": 0.0002490970974281599, | |
| "loss": 3.1499, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 42.7, | |
| "eval_accuracy": 0.4150287828947368, | |
| "eval_loss": 3.491703987121582, | |
| "eval_runtime": 128.345, | |
| "eval_samples_per_second": 187.425, | |
| "eval_steps_per_second": 5.859, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 42.9, | |
| "learning_rate": 0.00024781373367899597, | |
| "loss": 3.1374, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 43.12, | |
| "learning_rate": 0.00024652827840222606, | |
| "loss": 3.3333, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 43.33, | |
| "learning_rate": 0.00024524078782782807, | |
| "loss": 3.1338, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 43.53, | |
| "learning_rate": 0.00024395131827481062, | |
| "loss": 3.1092, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 43.74, | |
| "learning_rate": 0.0002426599261487494, | |
| "loss": 3.131, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 43.74, | |
| "eval_accuracy": 0.4168157298218596, | |
| "eval_loss": 3.4700751304626465, | |
| "eval_runtime": 128.1928, | |
| "eval_samples_per_second": 187.647, | |
| "eval_steps_per_second": 5.866, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 43.95, | |
| "learning_rate": 0.00024136666793931935, | |
| "loss": 3.1197, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 44.16, | |
| "learning_rate": 0.00024007160021782427, | |
| "loss": 3.3001, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 44.37, | |
| "learning_rate": 0.0002387747796347217, | |
| "loss": 3.0929, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 44.58, | |
| "learning_rate": 0.00023747626291714498, | |
| "loss": 3.0968, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 44.78, | |
| "learning_rate": 0.000236176106866422, | |
| "loss": 3.0942, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 44.78, | |
| "eval_accuracy": 0.4189149215354626, | |
| "eval_loss": 3.4530041217803955, | |
| "eval_runtime": 128.3801, | |
| "eval_samples_per_second": 187.373, | |
| "eval_steps_per_second": 5.858, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 44.99, | |
| "learning_rate": 0.00023487436835559035, | |
| "loss": 3.1072, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 45.21, | |
| "learning_rate": 0.00023357110432690954, | |
| "loss": 3.268, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 45.41, | |
| "learning_rate": 0.00023226637178937022, | |
| "loss": 3.0772, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 45.62, | |
| "learning_rate": 0.00023096022781620034, | |
| "loss": 3.071, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 45.82, | |
| "learning_rate": 0.0002296527295423684, | |
| "loss": 3.0683, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 45.82, | |
| "eval_accuracy": 0.42115190069686975, | |
| "eval_loss": 3.4319911003112793, | |
| "eval_runtime": 128.2347, | |
| "eval_samples_per_second": 187.586, | |
| "eval_steps_per_second": 5.864, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 46.04, | |
| "learning_rate": 0.00022834393416208486, | |
| "loss": 3.2606, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 46.25, | |
| "learning_rate": 0.0002270338989262994, | |
| "loss": 3.0464, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 46.45, | |
| "learning_rate": 0.00022572268114019726, | |
| "loss": 3.0424, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 46.66, | |
| "learning_rate": 0.00022441033816069202, | |
| "loss": 3.0469, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 46.86, | |
| "learning_rate": 0.00022309692739391727, | |
| "loss": 3.0363, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 46.86, | |
| "eval_accuracy": 0.42269065003604217, | |
| "eval_loss": 3.419463872909546, | |
| "eval_runtime": 128.2052, | |
| "eval_samples_per_second": 187.629, | |
| "eval_steps_per_second": 5.866, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 47.08, | |
| "learning_rate": 0.00022178250629271452, | |
| "loss": 3.2579, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 47.29, | |
| "learning_rate": 0.00022046713235412103, | |
| "loss": 3.0223, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 47.49, | |
| "learning_rate": 0.00021915086311685404, | |
| "loss": 3.0431, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 47.7, | |
| "learning_rate": 0.00021783375615879415, | |
| "loss": 3.0339, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 47.9, | |
| "learning_rate": 0.0002165158690944665, | |
| "loss": 3.0264, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 47.9, | |
| "eval_accuracy": 0.4248503269779865, | |
| "eval_loss": 3.4046127796173096, | |
| "eval_runtime": 128.0991, | |
| "eval_samples_per_second": 187.784, | |
| "eval_steps_per_second": 5.87, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 48.12, | |
| "learning_rate": 0.00021519725957252063, | |
| "loss": 3.2189, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 48.33, | |
| "learning_rate": 0.00021387798527320882, | |
| "loss": 3.0121, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 48.53, | |
| "learning_rate": 0.0002125581039058627, | |
| "loss": 3.0031, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 48.74, | |
| "learning_rate": 0.0002112376732063691, | |
| "loss": 2.9933, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 48.95, | |
| "learning_rate": 0.00020991675093464448, | |
| "loss": 3.0079, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 48.95, | |
| "eval_accuracy": 0.4266670495134685, | |
| "eval_loss": 3.3874006271362305, | |
| "eval_runtime": 128.1597, | |
| "eval_samples_per_second": 187.696, | |
| "eval_steps_per_second": 5.868, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 49.16, | |
| "learning_rate": 0.00020859539487210813, | |
| "loss": 3.2002, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 49.37, | |
| "learning_rate": 0.0002072736628191549, | |
| "loss": 2.9868, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 49.58, | |
| "learning_rate": 0.0002059516125926265, | |
| "loss": 2.9996, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 49.78, | |
| "learning_rate": 0.00020462930202328278, | |
| "loss": 2.9719, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 49.99, | |
| "learning_rate": 0.00020330678895327174, | |
| "loss": 2.9869, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 49.99, | |
| "eval_accuracy": 0.42770797209455824, | |
| "eval_loss": 3.3792383670806885, | |
| "eval_runtime": 128.195, | |
| "eval_samples_per_second": 187.644, | |
| "eval_steps_per_second": 5.866, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 50.21, | |
| "learning_rate": 0.00020198413123359926, | |
| "loss": 3.1735, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 50.41, | |
| "learning_rate": 0.00020066138672159903, | |
| "loss": 2.9707, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 50.62, | |
| "learning_rate": 0.00019933861327840098, | |
| "loss": 2.9682, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 50.82, | |
| "learning_rate": 0.00019801586876640073, | |
| "loss": 2.9752, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 51.04, | |
| "learning_rate": 0.0001966932110467283, | |
| "loss": 3.1592, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 51.04, | |
| "eval_accuracy": 0.4289155778595229, | |
| "eval_loss": 3.3654892444610596, | |
| "eval_runtime": 128.18, | |
| "eval_samples_per_second": 187.666, | |
| "eval_steps_per_second": 5.867, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 51.25, | |
| "learning_rate": 0.00019537069797671724, | |
| "loss": 2.948, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 51.45, | |
| "learning_rate": 0.0001940483874073735, | |
| "loss": 2.9368, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 51.66, | |
| "learning_rate": 0.00019272633718084517, | |
| "loss": 2.9466, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 51.86, | |
| "learning_rate": 0.0001914046051278919, | |
| "loss": 2.9541, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 52.08, | |
| "learning_rate": 0.00019008324906535554, | |
| "loss": 3.1353, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 52.08, | |
| "eval_accuracy": 0.43104300088517533, | |
| "eval_loss": 3.3548085689544678, | |
| "eval_runtime": 128.179, | |
| "eval_samples_per_second": 187.667, | |
| "eval_steps_per_second": 5.867, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 52.29, | |
| "learning_rate": 0.0001887623267936309, | |
| "loss": 2.9264, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 52.49, | |
| "learning_rate": 0.00018744189609413734, | |
| "loss": 2.9325, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 52.7, | |
| "learning_rate": 0.0001861220147267912, | |
| "loss": 2.9263, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 52.9, | |
| "learning_rate": 0.0001848027404274794, | |
| "loss": 2.9275, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 53.12, | |
| "learning_rate": 0.00018348413090553354, | |
| "loss": 3.1257, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 53.12, | |
| "eval_accuracy": 0.43083924373522625, | |
| "eval_loss": 3.348921775817871, | |
| "eval_runtime": 128.2056, | |
| "eval_samples_per_second": 187.628, | |
| "eval_steps_per_second": 5.866, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 53.33, | |
| "learning_rate": 0.00018216624384120595, | |
| "loss": 2.9018, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 53.53, | |
| "learning_rate": 0.00018084913688314597, | |
| "loss": 2.9135, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 53.74, | |
| "learning_rate": 0.000179532867645879, | |
| "loss": 2.9067, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 53.95, | |
| "learning_rate": 0.0001782174937072855, | |
| "loss": 2.9146, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 54.16, | |
| "learning_rate": 0.00017690307260608278, | |
| "loss": 3.0822, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 54.16, | |
| "eval_accuracy": 0.4326622137249495, | |
| "eval_loss": 3.3352506160736084, | |
| "eval_runtime": 128.2781, | |
| "eval_samples_per_second": 187.522, | |
| "eval_steps_per_second": 5.862, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 54.37, | |
| "learning_rate": 0.000175589661839308, | |
| "loss": 2.8995, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 54.58, | |
| "learning_rate": 0.00017427731885980282, | |
| "loss": 2.8945, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 54.78, | |
| "learning_rate": 0.0001729661010737007, | |
| "loss": 2.905, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 54.99, | |
| "learning_rate": 0.00017165606583791515, | |
| "loss": 2.9128, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 55.21, | |
| "learning_rate": 0.00017034727045763158, | |
| "loss": 3.0771, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 55.21, | |
| "eval_accuracy": 0.434098312415683, | |
| "eval_loss": 3.3219847679138184, | |
| "eval_runtime": 128.3146, | |
| "eval_samples_per_second": 187.469, | |
| "eval_steps_per_second": 5.861, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 55.41, | |
| "learning_rate": 0.00016903977218379974, | |
| "loss": 2.8695, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 55.62, | |
| "learning_rate": 0.00016773362821062983, | |
| "loss": 2.8839, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 55.82, | |
| "learning_rate": 0.00016642889567309048, | |
| "loss": 2.8887, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 56.04, | |
| "learning_rate": 0.0001651256316444097, | |
| "loss": 3.0754, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 56.25, | |
| "learning_rate": 0.0001638238931335781, | |
| "loss": 2.8639, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 56.25, | |
| "eval_accuracy": 0.4353990105725288, | |
| "eval_loss": 3.3119492530822754, | |
| "eval_runtime": 128.0745, | |
| "eval_samples_per_second": 187.82, | |
| "eval_steps_per_second": 5.872, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 56.45, | |
| "learning_rate": 0.00016252373708285504, | |
| "loss": 2.8653, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 56.66, | |
| "learning_rate": 0.00016122522036527838, | |
| "loss": 2.8696, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 56.86, | |
| "learning_rate": 0.00015992839978217578, | |
| "loss": 2.8665, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 57.08, | |
| "learning_rate": 0.00015863333206068067, | |
| "loss": 3.0651, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 57.29, | |
| "learning_rate": 0.00015734007385125067, | |
| "loss": 2.8477, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 57.29, | |
| "eval_accuracy": 0.4360402472560164, | |
| "eval_loss": 3.310389280319214, | |
| "eval_runtime": 128.2649, | |
| "eval_samples_per_second": 187.542, | |
| "eval_steps_per_second": 5.863, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 57.49, | |
| "learning_rate": 0.0001560486817251894, | |
| "loss": 2.8511, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 57.7, | |
| "learning_rate": 0.000154759212172172, | |
| "loss": 2.8615, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 57.9, | |
| "learning_rate": 0.00015347172159777396, | |
| "loss": 2.8619, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 58.12, | |
| "learning_rate": 0.000152186266321004, | |
| "loss": 3.0316, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 58.33, | |
| "learning_rate": 0.0001509029025718402, | |
| "loss": 2.8373, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 58.33, | |
| "eval_accuracy": 0.4378144877232535, | |
| "eval_loss": 3.295414686203003, | |
| "eval_runtime": 128.0673, | |
| "eval_samples_per_second": 187.831, | |
| "eval_steps_per_second": 5.872, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 58.53, | |
| "learning_rate": 0.0001496216864887704, | |
| "loss": 2.8292, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 58.74, | |
| "learning_rate": 0.00014834267411633674, | |
| "loss": 2.8361, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 58.95, | |
| "learning_rate": 0.0001470659214026837, | |
| "loss": 2.8417, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 59.16, | |
| "learning_rate": 0.00014579148419711119, | |
| "loss": 3.0263, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 59.37, | |
| "learning_rate": 0.00014451941824763113, | |
| "loss": 2.818, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 59.37, | |
| "eval_accuracy": 0.43805501144654146, | |
| "eval_loss": 3.2935194969177246, | |
| "eval_runtime": 128.0232, | |
| "eval_samples_per_second": 187.896, | |
| "eval_steps_per_second": 5.874, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 59.58, | |
| "learning_rate": 0.000143249779198529, | |
| "loss": 2.8253, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 59.78, | |
| "learning_rate": 0.00014198262258793002, | |
| "loss": 2.8424, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 59.99, | |
| "learning_rate": 0.00014071800384536927, | |
| "loss": 2.8335, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 60.21, | |
| "learning_rate": 0.00013945597828936737, | |
| "loss": 2.9887, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 60.41, | |
| "learning_rate": 0.00013819660112501054, | |
| "loss": 2.8137, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 60.41, | |
| "eval_accuracy": 0.4394361605808428, | |
| "eval_loss": 3.278566598892212, | |
| "eval_runtime": 128.078, | |
| "eval_samples_per_second": 187.815, | |
| "eval_steps_per_second": 5.871, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 60.62, | |
| "learning_rate": 0.00013693992744153572, | |
| "loss": 2.8271, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 60.82, | |
| "learning_rate": 0.00013568601220992097, | |
| "loss": 2.8286, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 61.04, | |
| "learning_rate": 0.00013443491028048045, | |
| "loss": 3.0135, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 61.25, | |
| "learning_rate": 0.0001331866763804658, | |
| "loss": 2.8038, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 61.45, | |
| "learning_rate": 0.0001319413651116714, | |
| "loss": 2.7985, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 61.45, | |
| "eval_accuracy": 0.4401244630436134, | |
| "eval_loss": 3.2746615409851074, | |
| "eval_runtime": 128.0922, | |
| "eval_samples_per_second": 187.794, | |
| "eval_steps_per_second": 5.871, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 61.66, | |
| "learning_rate": 0.00013069903094804644, | |
| "loss": 2.7993, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 61.86, | |
| "learning_rate": 0.0001294597282333118, | |
| "loss": 2.8132, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 62.08, | |
| "learning_rate": 0.00012822351117858303, | |
| "loss": 2.9785, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 62.29, | |
| "learning_rate": 0.0001269904338599989, | |
| "loss": 2.7959, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 62.49, | |
| "learning_rate": 0.0001257605502163558, | |
| "loss": 2.7936, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 62.49, | |
| "eval_accuracy": 0.44108544914689357, | |
| "eval_loss": 3.266845941543579, | |
| "eval_runtime": 128.1403, | |
| "eval_samples_per_second": 187.724, | |
| "eval_steps_per_second": 5.869, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 62.7, | |
| "learning_rate": 0.00012453391404674885, | |
| "loss": 2.7904, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 62.9, | |
| "learning_rate": 0.00012331057900821768, | |
| "loss": 2.7934, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 63.12, | |
| "learning_rate": 0.0001220905986134, | |
| "loss": 2.9571, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 63.33, | |
| "learning_rate": 0.00012087402622819039, | |
| "loss": 2.7925, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 63.53, | |
| "learning_rate": 0.00011966091506940616, | |
| "loss": 2.7764, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 63.53, | |
| "eval_accuracy": 0.441903341927903, | |
| "eval_loss": 3.256887197494507, | |
| "eval_runtime": 128.1402, | |
| "eval_samples_per_second": 187.724, | |
| "eval_steps_per_second": 5.869, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 63.74, | |
| "learning_rate": 0.00011845131820245934, | |
| "loss": 2.7851, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 63.95, | |
| "learning_rate": 0.00011724528853903536, | |
| "loss": 2.7837, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 64.16, | |
| "learning_rate": 0.00011604287883477889, | |
| "loss": 2.9344, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 64.37, | |
| "learning_rate": 0.00011484414168698547, | |
| "loss": 2.7703, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 64.58, | |
| "learning_rate": 0.00011364912953230145, | |
| "loss": 2.7819, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 64.58, | |
| "eval_accuracy": 0.44339571520227505, | |
| "eval_loss": 3.2492308616638184, | |
| "eval_runtime": 128.0362, | |
| "eval_samples_per_second": 187.877, | |
| "eval_steps_per_second": 5.873, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 64.78, | |
| "learning_rate": 0.00011245789464442964, | |
| "loss": 2.7841, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 64.99, | |
| "learning_rate": 0.00011127048913184326, | |
| "loss": 2.7794, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 65.21, | |
| "learning_rate": 0.00011008696493550599, | |
| "loss": 2.9422, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 65.41, | |
| "learning_rate": 0.00010890737382660015, | |
| "loss": 2.7573, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 65.62, | |
| "learning_rate": 0.00010773176740426248, | |
| "loss": 2.7672, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 65.62, | |
| "eval_accuracy": 0.4433201371393935, | |
| "eval_loss": 3.2493698596954346, | |
| "eval_runtime": 128.0217, | |
| "eval_samples_per_second": 187.898, | |
| "eval_steps_per_second": 5.874, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 65.82, | |
| "learning_rate": 0.00010656019709332606, | |
| "loss": 2.7557, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 66.04, | |
| "learning_rate": 0.00010539271414207186, | |
| "loss": 2.9353, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 66.25, | |
| "learning_rate": 0.00010422936961998609, | |
| "loss": 2.7494, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 66.45, | |
| "learning_rate": 0.00010307021441552707, | |
| "loss": 2.7401, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 66.66, | |
| "learning_rate": 0.00010191529923389845, | |
| "loss": 2.7629, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 66.66, | |
| "eval_accuracy": 0.44430680533611233, | |
| "eval_loss": 3.240968704223633, | |
| "eval_runtime": 128.0927, | |
| "eval_samples_per_second": 187.794, | |
| "eval_steps_per_second": 5.871, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 66.86, | |
| "learning_rate": 0.00010076467459483155, | |
| "loss": 2.7537, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 67.08, | |
| "learning_rate": 9.961839083037592e-05, | |
| "loss": 2.9359, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 67.29, | |
| "learning_rate": 9.847649808269658e-05, | |
| "loss": 2.7575, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 67.49, | |
| "learning_rate": 9.733904630188176e-05, | |
| "loss": 2.7294, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 67.7, | |
| "learning_rate": 9.620608524375703e-05, | |
| "loss": 2.747, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 67.7, | |
| "eval_accuracy": 0.4446199448310505, | |
| "eval_loss": 3.236819267272949, | |
| "eval_runtime": 127.9991, | |
| "eval_samples_per_second": 187.931, | |
| "eval_steps_per_second": 5.875, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 67.9, | |
| "learning_rate": 9.507766446770934e-05, | |
| "loss": 2.7458, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 68.12, | |
| "learning_rate": 9.39538333345191e-05, | |
| "loss": 2.9246, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 68.33, | |
| "learning_rate": 9.283464100420063e-05, | |
| "loss": 2.741, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 68.53, | |
| "learning_rate": 9.17201364338524e-05, | |
| "loss": 2.7421, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 68.74, | |
| "learning_rate": 9.061036837551466e-05, | |
| "loss": 2.7303, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 68.74, | |
| "eval_accuracy": 0.44596990309042184, | |
| "eval_loss": 3.224606990814209, | |
| "eval_runtime": 128.0175, | |
| "eval_samples_per_second": 187.904, | |
| "eval_steps_per_second": 5.874, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 68.95, | |
| "learning_rate": 8.950538537403736e-05, | |
| "loss": 2.7291, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 69.16, | |
| "learning_rate": 8.840523576495681e-05, | |
| "loss": 2.903, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 69.37, | |
| "learning_rate": 8.730996767238072e-05, | |
| "loss": 2.7319, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 69.58, | |
| "learning_rate": 8.621962900688378e-05, | |
| "loss": 2.7166, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 69.78, | |
| "learning_rate": 8.513426746341128e-05, | |
| "loss": 2.7461, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 69.78, | |
| "eval_accuracy": 0.44624793300809595, | |
| "eval_loss": 3.2212436199188232, | |
| "eval_runtime": 128.0443, | |
| "eval_samples_per_second": 187.865, | |
| "eval_steps_per_second": 5.873, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 69.99, | |
| "learning_rate": 8.405393051919333e-05, | |
| "loss": 2.7214, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 70.21, | |
| "learning_rate": 8.29786654316677e-05, | |
| "loss": 2.8969, | |
| "step": 3370 | |
| }, | |
| { | |
| "epoch": 70.41, | |
| "learning_rate": 8.190851923641259e-05, | |
| "loss": 2.6964, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 70.62, | |
| "learning_rate": 8.084353874508947e-05, | |
| "loss": 2.7295, | |
| "step": 3390 | |
| }, | |
| { | |
| "epoch": 70.82, | |
| "learning_rate": 7.978377054339499e-05, | |
| "loss": 2.7179, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 70.82, | |
| "eval_accuracy": 0.4470331759822518, | |
| "eval_loss": 3.221658706665039, | |
| "eval_runtime": 127.9471, | |
| "eval_samples_per_second": 188.007, | |
| "eval_steps_per_second": 5.877, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 71.04, | |
| "learning_rate": 7.872926098902358e-05, | |
| "loss": 2.9027, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 71.25, | |
| "learning_rate": 7.768005620963916e-05, | |
| "loss": 2.7053, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 71.45, | |
| "learning_rate": 7.663620210085781e-05, | |
| "loss": 2.709, | |
| "step": 3430 | |
| }, | |
| { | |
| "epoch": 71.66, | |
| "learning_rate": 7.55977443242399e-05, | |
| "loss": 2.7125, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 71.86, | |
| "learning_rate": 7.456472830529259e-05, | |
| "loss": 2.7184, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 71.86, | |
| "eval_accuracy": 0.44788752382659924, | |
| "eval_loss": 3.213238000869751, | |
| "eval_runtime": 127.9515, | |
| "eval_samples_per_second": 188.001, | |
| "eval_steps_per_second": 5.877, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 72.08, | |
| "learning_rate": 7.353719923148324e-05, | |
| "loss": 2.8953, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 72.29, | |
| "learning_rate": 7.251520205026205e-05, | |
| "loss": 2.6971, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 72.49, | |
| "learning_rate": 7.149878146709676e-05, | |
| "loss": 2.6983, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 72.7, | |
| "learning_rate": 7.048798194351625e-05, | |
| "loss": 2.7034, | |
| "step": 3490 | |
| }, | |
| { | |
| "epoch": 72.9, | |
| "learning_rate": 6.948284769516627e-05, | |
| "loss": 2.7077, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 72.9, | |
| "eval_accuracy": 0.44867082596467595, | |
| "eval_loss": 3.208606243133545, | |
| "eval_runtime": 128.1176, | |
| "eval_samples_per_second": 187.757, | |
| "eval_steps_per_second": 5.87, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 73.12, | |
| "learning_rate": 6.848342268987511e-05, | |
| "loss": 2.8784, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 73.33, | |
| "learning_rate": 6.748975064573007e-05, | |
| "loss": 2.694, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 73.53, | |
| "learning_rate": 6.650187502916552e-05, | |
| "loss": 2.6991, | |
| "step": 3530 | |
| }, | |
| { | |
| "epoch": 73.74, | |
| "learning_rate": 6.551983905306107e-05, | |
| "loss": 2.7075, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 73.95, | |
| "learning_rate": 6.454368567485183e-05, | |
| "loss": 2.6916, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 73.95, | |
| "eval_accuracy": 0.44818311301861347, | |
| "eval_loss": 3.2057085037231445, | |
| "eval_runtime": 128.0769, | |
| "eval_samples_per_second": 187.817, | |
| "eval_steps_per_second": 5.871, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 74.16, | |
| "learning_rate": 6.35734575946487e-05, | |
| "loss": 2.884, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 74.37, | |
| "learning_rate": 6.260919725337109e-05, | |
| "loss": 2.6885, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 74.58, | |
| "learning_rate": 6.165094683089015e-05, | |
| "loss": 2.7009, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 74.78, | |
| "learning_rate": 6.069874824418356e-05, | |
| "loss": 2.6924, | |
| "step": 3590 | |
| }, | |
| { | |
| "epoch": 74.99, | |
| "learning_rate": 5.975264314550229e-05, | |
| "loss": 2.6934, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 74.99, | |
| "eval_accuracy": 0.44951231576252726, | |
| "eval_loss": 3.201040506362915, | |
| "eval_runtime": 128.0178, | |
| "eval_samples_per_second": 187.904, | |
| "eval_steps_per_second": 5.874, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 75.21, | |
| "learning_rate": 5.881267292054828e-05, | |
| "loss": 2.8607, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 75.41, | |
| "learning_rate": 5.787887868666417e-05, | |
| "loss": 2.678, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 75.62, | |
| "learning_rate": 5.6951301291034945e-05, | |
| "loss": 2.696, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 75.82, | |
| "learning_rate": 5.602998130890065e-05, | |
| "loss": 2.6944, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 76.04, | |
| "learning_rate": 5.511495904178221e-05, | |
| "loss": 2.8585, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 76.04, | |
| "eval_accuracy": 0.44973287373334114, | |
| "eval_loss": 3.1979689598083496, | |
| "eval_runtime": 127.9186, | |
| "eval_samples_per_second": 188.049, | |
| "eval_steps_per_second": 5.879, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 76.25, | |
| "learning_rate": 5.4206274515717736e-05, | |
| "loss": 2.6924, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 76.45, | |
| "learning_rate": 5.330396747951205e-05, | |
| "loss": 2.6796, | |
| "step": 3670 | |
| }, | |
| { | |
| "epoch": 76.66, | |
| "learning_rate": 5.240807740299811e-05, | |
| "loss": 2.684, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 76.86, | |
| "learning_rate": 5.1518643475310034e-05, | |
| "loss": 2.6842, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 77.08, | |
| "learning_rate": 5.0635704603169287e-05, | |
| "loss": 2.8559, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 77.08, | |
| "eval_accuracy": 0.4502848474176814, | |
| "eval_loss": 3.1939539909362793, | |
| "eval_runtime": 128.1047, | |
| "eval_samples_per_second": 187.776, | |
| "eval_steps_per_second": 5.87, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 77.29, | |
| "learning_rate": 4.975929940918236e-05, | |
| "loss": 2.6777, | |
| "step": 3710 | |
| }, | |
| { | |
| "epoch": 77.49, | |
| "learning_rate": 4.8889466230151646e-05, | |
| "loss": 2.6673, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 77.7, | |
| "learning_rate": 4.8026243115398314e-05, | |
| "loss": 2.6694, | |
| "step": 3730 | |
| }, | |
| { | |
| "epoch": 77.9, | |
| "learning_rate": 4.7169667825097775e-05, | |
| "loss": 2.6734, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 78.12, | |
| "learning_rate": 4.631977782862824e-05, | |
| "loss": 2.8519, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 78.12, | |
| "eval_accuracy": 0.4506198615318044, | |
| "eval_loss": 3.1939969062805176, | |
| "eval_runtime": 128.0925, | |
| "eval_samples_per_second": 187.794, | |
| "eval_steps_per_second": 5.871, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 78.33, | |
| "learning_rate": 4.547661030293129e-05, | |
| "loss": 2.6742, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 78.53, | |
| "learning_rate": 4.464020213088611e-05, | |
| "loss": 2.6767, | |
| "step": 3770 | |
| }, | |
| { | |
| "epoch": 78.74, | |
| "learning_rate": 4.381058989969564e-05, | |
| "loss": 2.6641, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 78.95, | |
| "learning_rate": 4.298780989928646e-05, | |
| "loss": 2.6726, | |
| "step": 3790 | |
| }, | |
| { | |
| "epoch": 79.16, | |
| "learning_rate": 4.217189812072131e-05, | |
| "loss": 2.8391, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 79.16, | |
| "eval_accuracy": 0.4509423513828209, | |
| "eval_loss": 3.1897408962249756, | |
| "eval_runtime": 127.8217, | |
| "eval_samples_per_second": 188.192, | |
| "eval_steps_per_second": 5.883, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 79.37, | |
| "learning_rate": 4.136289025462443e-05, | |
| "loss": 2.6616, | |
| "step": 3810 | |
| }, | |
| { | |
| "epoch": 79.58, | |
| "learning_rate": 4.0560821689620856e-05, | |
| "loss": 2.6701, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 79.78, | |
| "learning_rate": 3.976572751078782e-05, | |
| "loss": 2.6546, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 79.99, | |
| "learning_rate": 3.8977642498120594e-05, | |
| "loss": 2.6719, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 80.21, | |
| "learning_rate": 3.819660112501053e-05, | |
| "loss": 2.845, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 80.21, | |
| "eval_accuracy": 0.45101718878618524, | |
| "eval_loss": 3.1857643127441406, | |
| "eval_runtime": 127.9876, | |
| "eval_samples_per_second": 187.948, | |
| "eval_steps_per_second": 5.876, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 80.41, | |
| "learning_rate": 3.742263755673758e-05, | |
| "loss": 2.6657, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 80.62, | |
| "learning_rate": 3.6655785648975585e-05, | |
| "loss": 2.6601, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 80.82, | |
| "learning_rate": 3.589607894631111e-05, | |
| "loss": 2.6666, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 81.04, | |
| "learning_rate": 3.514355068077655e-05, | |
| "loss": 2.8323, | |
| "step": 3890 | |
| }, | |
| { | |
| "epoch": 81.25, | |
| "learning_rate": 3.439823377039599e-05, | |
| "loss": 2.6636, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 81.25, | |
| "eval_accuracy": 0.45183219751680725, | |
| "eval_loss": 3.1818630695343018, | |
| "eval_runtime": 128.0672, | |
| "eval_samples_per_second": 187.831, | |
| "eval_steps_per_second": 5.872, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 81.45, | |
| "learning_rate": 3.36601608177457e-05, | |
| "loss": 2.6586, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 81.66, | |
| "learning_rate": 3.292936410852754e-05, | |
| "loss": 2.6674, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 81.86, | |
| "learning_rate": 3.220587561015709e-05, | |
| "loss": 2.6689, | |
| "step": 3930 | |
| }, | |
| { | |
| "epoch": 82.08, | |
| "learning_rate": 3.148972697036507e-05, | |
| "loss": 2.8232, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 82.29, | |
| "learning_rate": 3.078094951581289e-05, | |
| "loss": 2.6569, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 82.29, | |
| "eval_accuracy": 0.4517055966540888, | |
| "eval_loss": 3.183380603790283, | |
| "eval_runtime": 128.002, | |
| "eval_samples_per_second": 187.927, | |
| "eval_steps_per_second": 5.875, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 82.49, | |
| "learning_rate": 3.007957425072265e-05, | |
| "loss": 2.6544, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 82.7, | |
| "learning_rate": 2.9385631855520546e-05, | |
| "loss": 2.6622, | |
| "step": 3970 | |
| }, | |
| { | |
| "epoch": 82.9, | |
| "learning_rate": 2.8699152685494925e-05, | |
| "loss": 2.6505, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 83.12, | |
| "learning_rate": 2.8020166769468616e-05, | |
| "loss": 2.8267, | |
| "step": 3990 | |
| }, | |
| { | |
| "epoch": 83.33, | |
| "learning_rate": 2.7348703808485223e-05, | |
| "loss": 2.647, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 83.33, | |
| "eval_accuracy": 0.45166349665740935, | |
| "eval_loss": 3.1797752380371094, | |
| "eval_runtime": 128.0064, | |
| "eval_samples_per_second": 187.92, | |
| "eval_steps_per_second": 5.875, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 83.53, | |
| "learning_rate": 2.6684793174509915e-05, | |
| "loss": 2.6432, | |
| "step": 4010 | |
| }, | |
| { | |
| "epoch": 83.74, | |
| "learning_rate": 2.6028463909144574e-05, | |
| "loss": 2.6626, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 83.95, | |
| "learning_rate": 2.5379744722357403e-05, | |
| "loss": 2.6586, | |
| "step": 4030 | |
| }, | |
| { | |
| "epoch": 84.16, | |
| "learning_rate": 2.473866399122733e-05, | |
| "loss": 2.8349, | |
| "step": 4040 | |
| }, | |
| { | |
| "epoch": 84.37, | |
| "learning_rate": 2.410524975870221e-05, | |
| "loss": 2.6665, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 84.37, | |
| "eval_accuracy": 0.45251206379734554, | |
| "eval_loss": 3.178643226623535, | |
| "eval_runtime": 127.9924, | |
| "eval_samples_per_second": 187.941, | |
| "eval_steps_per_second": 5.875, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 84.58, | |
| "learning_rate": 2.347952973237262e-05, | |
| "loss": 2.6462, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 84.78, | |
| "learning_rate": 2.286153128325954e-05, | |
| "loss": 2.6444, | |
| "step": 4070 | |
| }, | |
| { | |
| "epoch": 84.99, | |
| "learning_rate": 2.2251281444617257e-05, | |
| "loss": 2.6442, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 85.21, | |
| "learning_rate": 2.1648806910750575e-05, | |
| "loss": 2.8258, | |
| "step": 4090 | |
| }, | |
| { | |
| "epoch": 85.41, | |
| "learning_rate": 2.1054134035847307e-05, | |
| "loss": 2.6382, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 85.41, | |
| "eval_accuracy": 0.4524524692406656, | |
| "eval_loss": 3.173250198364258, | |
| "eval_runtime": 127.972, | |
| "eval_samples_per_second": 187.971, | |
| "eval_steps_per_second": 5.876, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 85.62, | |
| "learning_rate": 2.0467288832825583e-05, | |
| "loss": 2.6655, | |
| "step": 4110 | |
| }, | |
| { | |
| "epoch": 85.82, | |
| "learning_rate": 1.9888296972195587e-05, | |
| "loss": 2.6459, | |
| "step": 4120 | |
| }, | |
| { | |
| "epoch": 86.04, | |
| "learning_rate": 1.931718378093703e-05, | |
| "loss": 2.8333, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 86.25, | |
| "learning_rate": 1.875397424139109e-05, | |
| "loss": 2.6533, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 86.45, | |
| "learning_rate": 1.81986929901675e-05, | |
| "loss": 2.6346, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 86.45, | |
| "eval_accuracy": 0.4532207710219251, | |
| "eval_loss": 3.1699652671813965, | |
| "eval_runtime": 128.0545, | |
| "eval_samples_per_second": 187.85, | |
| "eval_steps_per_second": 5.872, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 86.66, | |
| "learning_rate": 1.765136431706711e-05, | |
| "loss": 2.6558, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 86.86, | |
| "learning_rate": 1.711201216401912e-05, | |
| "loss": 2.6422, | |
| "step": 4170 | |
| }, | |
| { | |
| "epoch": 87.08, | |
| "learning_rate": 1.6580660124034032e-05, | |
| "loss": 2.8243, | |
| "step": 4180 | |
| }, | |
| { | |
| "epoch": 87.29, | |
| "learning_rate": 1.605733144017132e-05, | |
| "loss": 2.6443, | |
| "step": 4190 | |
| }, | |
| { | |
| "epoch": 87.49, | |
| "learning_rate": 1.5542049004523053e-05, | |
| "loss": 2.6457, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 87.49, | |
| "eval_accuracy": 0.45290158205139586, | |
| "eval_loss": 3.1713671684265137, | |
| "eval_runtime": 127.871, | |
| "eval_samples_per_second": 188.119, | |
| "eval_steps_per_second": 5.881, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 87.7, | |
| "learning_rate": 1.503483535721224e-05, | |
| "loss": 2.6578, | |
| "step": 4210 | |
| }, | |
| { | |
| "epoch": 87.9, | |
| "learning_rate": 1.4535712685406921e-05, | |
| "loss": 2.646, | |
| "step": 4220 | |
| }, | |
| { | |
| "epoch": 88.12, | |
| "learning_rate": 1.4044702822349731e-05, | |
| "loss": 2.8075, | |
| "step": 4230 | |
| }, | |
| { | |
| "epoch": 88.33, | |
| "learning_rate": 1.3561827246402692e-05, | |
| "loss": 2.6405, | |
| "step": 4240 | |
| }, | |
| { | |
| "epoch": 88.53, | |
| "learning_rate": 1.3087107080107853e-05, | |
| "loss": 2.6328, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 88.53, | |
| "eval_accuracy": 0.45369892675477164, | |
| "eval_loss": 3.168638229370117, | |
| "eval_runtime": 128.0728, | |
| "eval_samples_per_second": 187.823, | |
| "eval_steps_per_second": 5.872, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 88.74, | |
| "learning_rate": 1.2620563089263093e-05, | |
| "loss": 2.6377, | |
| "step": 4260 | |
| }, | |
| { | |
| "epoch": 88.95, | |
| "learning_rate": 1.2162215682014012e-05, | |
| "loss": 2.6645, | |
| "step": 4270 | |
| }, | |
| { | |
| "epoch": 89.16, | |
| "learning_rate": 1.1712084907961053e-05, | |
| "loss": 2.8112, | |
| "step": 4280 | |
| }, | |
| { | |
| "epoch": 89.37, | |
| "learning_rate": 1.127019045728246e-05, | |
| "loss": 2.6445, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 89.58, | |
| "learning_rate": 1.0836551659873074e-05, | |
| "loss": 2.6429, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 89.58, | |
| "eval_accuracy": 0.4534057770055075, | |
| "eval_loss": 3.171478033065796, | |
| "eval_runtime": 128.1308, | |
| "eval_samples_per_second": 187.738, | |
| "eval_steps_per_second": 5.869, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 89.78, | |
| "learning_rate": 1.0411187484498652e-05, | |
| "loss": 2.6458, | |
| "step": 4310 | |
| }, | |
| { | |
| "epoch": 89.99, | |
| "learning_rate": 9.99411653796627e-06, | |
| "loss": 2.6433, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 90.21, | |
| "learning_rate": 9.58535706431023e-06, | |
| "loss": 2.8107, | |
| "step": 4330 | |
| }, | |
| { | |
| "epoch": 90.41, | |
| "learning_rate": 9.184926943994044e-06, | |
| "loss": 2.6428, | |
| "step": 4340 | |
| }, | |
| { | |
| "epoch": 90.62, | |
| "learning_rate": 8.792843693128471e-06, | |
| "loss": 2.6369, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 90.62, | |
| "eval_accuracy": 0.4537699711828984, | |
| "eval_loss": 3.1687278747558594, | |
| "eval_runtime": 127.8724, | |
| "eval_samples_per_second": 188.117, | |
| "eval_steps_per_second": 5.881, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 90.82, | |
| "learning_rate": 8.409124462705032e-06, | |
| "loss": 2.6513, | |
| "step": 4360 | |
| }, | |
| { | |
| "epoch": 91.04, | |
| "learning_rate": 8.033786037845992e-06, | |
| "loss": 2.8, | |
| "step": 4370 | |
| }, | |
| { | |
| "epoch": 91.25, | |
| "learning_rate": 7.66684483706992e-06, | |
| "loss": 2.6456, | |
| "step": 4380 | |
| }, | |
| { | |
| "epoch": 91.45, | |
| "learning_rate": 7.308316911573721e-06, | |
| "loss": 2.6429, | |
| "step": 4390 | |
| }, | |
| { | |
| "epoch": 91.66, | |
| "learning_rate": 6.958217944530287e-06, | |
| "loss": 2.628, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 91.66, | |
| "eval_accuracy": 0.4539267525748088, | |
| "eval_loss": 3.165127754211426, | |
| "eval_runtime": 128.0531, | |
| "eval_samples_per_second": 187.852, | |
| "eval_steps_per_second": 5.873, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 91.86, | |
| "learning_rate": 6.616563250402585e-06, | |
| "loss": 2.6337, | |
| "step": 4410 | |
| }, | |
| { | |
| "epoch": 92.08, | |
| "learning_rate": 6.283367774273785e-06, | |
| "loss": 2.8133, | |
| "step": 4420 | |
| }, | |
| { | |
| "epoch": 92.29, | |
| "learning_rate": 5.958646091193387e-06, | |
| "loss": 2.6318, | |
| "step": 4430 | |
| }, | |
| { | |
| "epoch": 92.49, | |
| "learning_rate": 5.642412405539798e-06, | |
| "loss": 2.6365, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 92.7, | |
| "learning_rate": 5.334680550398852e-06, | |
| "loss": 2.6373, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 92.7, | |
| "eval_accuracy": 0.4538699774217477, | |
| "eval_loss": 3.1659765243530273, | |
| "eval_runtime": 128.2044, | |
| "eval_samples_per_second": 187.63, | |
| "eval_steps_per_second": 5.866, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 92.9, | |
| "learning_rate": 5.0354639869588e-06, | |
| "loss": 2.6355, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 93.12, | |
| "learning_rate": 4.744775803921475e-06, | |
| "loss": 2.8102, | |
| "step": 4470 | |
| }, | |
| { | |
| "epoch": 93.33, | |
| "learning_rate": 4.4626287169296846e-06, | |
| "loss": 2.6362, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 93.53, | |
| "learning_rate": 4.189035068011071e-06, | |
| "loss": 2.6226, | |
| "step": 4490 | |
| }, | |
| { | |
| "epoch": 93.74, | |
| "learning_rate": 3.924006825038129e-06, | |
| "loss": 2.6357, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 93.74, | |
| "eval_accuracy": 0.4537227251216398, | |
| "eval_loss": 3.1661999225616455, | |
| "eval_runtime": 128.0693, | |
| "eval_samples_per_second": 187.828, | |
| "eval_steps_per_second": 5.872, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 93.95, | |
| "learning_rate": 3.6675555812047956e-06, | |
| "loss": 2.6477, | |
| "step": 4510 | |
| }, | |
| { | |
| "epoch": 94.16, | |
| "learning_rate": 3.4196925545192604e-06, | |
| "loss": 2.808, | |
| "step": 4520 | |
| }, | |
| { | |
| "epoch": 94.37, | |
| "learning_rate": 3.1804285873132668e-06, | |
| "loss": 2.6339, | |
| "step": 4530 | |
| }, | |
| { | |
| "epoch": 94.58, | |
| "learning_rate": 2.9497741457678695e-06, | |
| "loss": 2.6228, | |
| "step": 4540 | |
| }, | |
| { | |
| "epoch": 94.78, | |
| "learning_rate": 2.7277393194555358e-06, | |
| "loss": 2.6302, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 94.78, | |
| "eval_accuracy": 0.45330473365422363, | |
| "eval_loss": 3.1695384979248047, | |
| "eval_runtime": 128.0386, | |
| "eval_samples_per_second": 187.873, | |
| "eval_steps_per_second": 5.873, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 94.78, | |
| "step": 4550, | |
| "total_flos": 1.602912650550436e+17, | |
| "train_loss": 3.832834072322636, | |
| "train_runtime": 39205.3218, | |
| "train_samples_per_second": 63.537, | |
| "train_steps_per_second": 0.122 | |
| } | |
| ], | |
| "max_steps": 4800, | |
| "num_train_epochs": 100, | |
| "total_flos": 1.602912650550436e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |