Buckets:
| {"kind": "eval", "optimizer_step": 0, "real_loss": 1.508930926521619, "shuffled_loss": 1.5424310167630513, "conditioning_gap": 0.03350009024143219, "mean_cosine": 0.24553453673919046, "fve_rawvar": -1.1008279928826261, "elapsed_seconds": 0.0} | |
| {"kind": "train", "optimizer_step": 1, "micro_step": 1, "loss": 1.4952398538589478, "mean_cosine": 0.2523800730705261, "fve_rawvar": -1.0670388704586937, "learning_rate": 6.12e-07, "grad_norm": 14.210945129394531, "pred_norm_raw": 100.62603759765625, "gold_norm_raw": 122.20645904541016, "backbone_norm_raw": 100.62603759765625, "elapsed_seconds": 28.106062650680542, "seconds_per_optimizer_step": 28.106062650680542} | |
| {"kind": "train", "optimizer_step": 2, "micro_step": 2, "loss": 1.5233330726623535, "mean_cosine": 0.23833346366882324, "fve_rawvar": -1.1058752986832845, "learning_rate": 1.224e-06, "grad_norm": 13.932377815246582, "pred_norm_raw": 99.6658935546875, "gold_norm_raw": 123.4200210571289, "backbone_norm_raw": 99.6658935546875, "elapsed_seconds": 53.561604738235474, "seconds_per_optimizer_step": 26.780802369117737} | |
| {"kind": "train", "optimizer_step": 3, "micro_step": 3, "loss": 1.4648756980895996, "mean_cosine": 0.2675621509552002, "fve_rawvar": -1.0250630696653142, "learning_rate": 1.8359999999999999e-06, "grad_norm": 12.80424976348877, "pred_norm_raw": 102.5604248046875, "gold_norm_raw": 118.54206085205078, "backbone_norm_raw": 102.59591674804688, "elapsed_seconds": 79.0200138092041, "seconds_per_optimizer_step": 26.340004603068035} | |
| {"kind": "train", "optimizer_step": 4, "micro_step": 4, "loss": 1.322189211845398, "mean_cosine": 0.338905394077301, "fve_rawvar": -0.8278114296727397, "learning_rate": 2.448e-06, "grad_norm": 8.716591835021973, "pred_norm_raw": 99.91044616699219, "gold_norm_raw": 118.4545669555664, "backbone_norm_raw": 99.9956283569336, "elapsed_seconds": 104.14414262771606, "seconds_per_optimizer_step": 26.036035656929016} | |
| {"kind": "train", "optimizer_step": 5, "micro_step": 5, "loss": 1.4200876951217651, "mean_cosine": 0.28995615243911743, "fve_rawvar": -0.9631475563610075, "learning_rate": 3.06e-06, "grad_norm": 4.52089786529541, "pred_norm_raw": 105.12814331054688, "gold_norm_raw": 123.75477600097656, "backbone_norm_raw": 105.24581909179688, "elapsed_seconds": 129.52023839950562, "seconds_per_optimizer_step": 25.904047679901122} | |
| {"kind": "train", "optimizer_step": 6, "micro_step": 6, "loss": 1.3236756324768066, "mean_cosine": 0.3381621837615967, "fve_rawvar": -0.8298662767362688, "learning_rate": 3.6719999999999997e-06, "grad_norm": 2.8692686557769775, "pred_norm_raw": 103.4139404296875, "gold_norm_raw": 123.82331848144531, "backbone_norm_raw": 103.50650024414062, "elapsed_seconds": 154.9790551662445, "seconds_per_optimizer_step": 25.829842527707417} | |
| {"kind": "train", "optimizer_step": 7, "micro_step": 7, "loss": 1.1955609321594238, "mean_cosine": 0.4022195339202881, "fve_rawvar": -0.6527588616618587, "learning_rate": 4.284e-06, "grad_norm": 3.236023187637329, "pred_norm_raw": 101.28021240234375, "gold_norm_raw": 122.87745666503906, "backbone_norm_raw": 101.34043884277344, "elapsed_seconds": 180.30031967163086, "seconds_per_optimizer_step": 25.757188524518693} | |
| {"kind": "train", "optimizer_step": 8, "micro_step": 8, "loss": 1.08029043674469, "mean_cosine": 0.45985478162765503, "fve_rawvar": -0.4934074412028888, "learning_rate": 4.896e-06, "grad_norm": 2.508291721343994, "pred_norm_raw": 102.86334991455078, "gold_norm_raw": 118.05038452148438, "backbone_norm_raw": 102.90579223632812, "elapsed_seconds": 205.58755779266357, "seconds_per_optimizer_step": 25.698444724082947} | |
| {"kind": "train", "optimizer_step": 9, "micro_step": 9, "loss": 0.9961938261985779, "mean_cosine": 0.5019030869007111, "fve_rawvar": -0.3771512014939129, "learning_rate": 5.508e-06, "grad_norm": 2.207770347595215, "pred_norm_raw": 106.35910034179688, "gold_norm_raw": 123.55357360839844, "backbone_norm_raw": 106.40802001953125, "elapsed_seconds": 230.83675599098206, "seconds_per_optimizer_step": 25.64852844344245} | |
| {"kind": "train", "optimizer_step": 10, "micro_step": 10, "loss": 0.9072206020355225, "mean_cosine": 0.5463896989822388, "fve_rawvar": -0.25415346818682605, "learning_rate": 6.12e-06, "grad_norm": 2.1245710849761963, "pred_norm_raw": 102.55419921875, "gold_norm_raw": 123.77955627441406, "backbone_norm_raw": 102.59397888183594, "elapsed_seconds": 256.20221734046936, "seconds_per_optimizer_step": 25.620221734046936} | |
| {"kind": "train", "optimizer_step": 11, "micro_step": 11, "loss": 1.0071208477020264, "mean_cosine": 0.4964395761489868, "fve_rawvar": -0.39225685703651636, "learning_rate": 6.1196568227316276e-06, "grad_norm": 1.9999390840530396, "pred_norm_raw": 102.53925323486328, "gold_norm_raw": 120.81561279296875, "backbone_norm_raw": 102.5723876953125, "elapsed_seconds": 281.4521384239197, "seconds_per_optimizer_step": 25.586558038538154} | |
| {"kind": "train", "optimizer_step": 12, "micro_step": 12, "loss": 0.954774022102356, "mean_cosine": 0.522612988948822, "fve_rawvar": -0.319891929777262, "learning_rate": 6.118627376453482e-06, "grad_norm": 2.3151137828826904, "pred_norm_raw": 100.20844268798828, "gold_norm_raw": 120.58914184570312, "backbone_norm_raw": 100.2379379272461, "elapsed_seconds": 306.6227331161499, "seconds_per_optimizer_step": 25.551894426345825} | |
| {"kind": "train", "optimizer_step": 13, "micro_step": 13, "loss": 0.7913830280303955, "mean_cosine": 0.6043084859848022, "fve_rawvar": -0.09401811096619084, "learning_rate": 6.116911917725157e-06, "grad_norm": 2.6185848712921143, "pred_norm_raw": 97.13922119140625, "gold_norm_raw": 113.90424346923828, "backbone_norm_raw": 97.17328643798828, "elapsed_seconds": 331.6671664714813, "seconds_per_optimizer_step": 25.51285895934472} | |
| {"kind": "train", "optimizer_step": 14, "micro_step": 14, "loss": 0.9318819046020508, "mean_cosine": 0.5340590476989746, "fve_rawvar": -0.2882455711157288, "learning_rate": 6.114510874074933e-06, "grad_norm": 2.869332790374756, "pred_norm_raw": 99.58211517333984, "gold_norm_raw": 119.07604217529297, "backbone_norm_raw": 99.60940551757812, "elapsed_seconds": 356.98475980758667, "seconds_per_optimizer_step": 25.49891141482762} | |
| {"kind": "train", "optimizer_step": 15, "micro_step": 15, "loss": 0.8584443926811218, "mean_cosine": 0.5707778036594391, "fve_rawvar": -0.1867246069048234, "learning_rate": 6.1114248438932226e-06, "grad_norm": 2.5222790241241455, "pred_norm_raw": 101.62621307373047, "gold_norm_raw": 124.57113647460938, "backbone_norm_raw": 101.65703582763672, "elapsed_seconds": 382.40016913414, "seconds_per_optimizer_step": 25.493344608942667} | |
| {"kind": "train", "optimizer_step": 16, "micro_step": 16, "loss": 0.715801477432251, "mean_cosine": 0.6420992612838745, "fve_rawvar": 0.010466799981509789, "learning_rate": 6.107654596283446e-06, "grad_norm": 2.450528860092163, "pred_norm_raw": 98.95343017578125, "gold_norm_raw": 129.8714599609375, "backbone_norm_raw": 98.96800231933594, "elapsed_seconds": 407.74107456207275, "seconds_per_optimizer_step": 25.483817160129547} | |
| {"kind": "train", "optimizer_step": 17, "micro_step": 17, "loss": 0.812847912311554, "mean_cosine": 0.593576043844223, "fve_rawvar": -0.12369144400673582, "learning_rate": 6.1032010708703485e-06, "grad_norm": 2.5101025104522705, "pred_norm_raw": 103.13805389404297, "gold_norm_raw": 123.32633972167969, "backbone_norm_raw": 103.15553283691406, "elapsed_seconds": 433.0550503730774, "seconds_per_optimizer_step": 25.473826492533963} | |
| {"kind": "train", "optimizer_step": 18, "micro_step": 18, "loss": 0.7900981903076172, "mean_cosine": 0.6049509048461914, "fve_rawvar": -0.09224193471703579, "learning_rate": 6.098065377565832e-06, "grad_norm": 3.058978796005249, "pred_norm_raw": 102.20268249511719, "gold_norm_raw": 122.515869140625, "backbone_norm_raw": 102.22065734863281, "elapsed_seconds": 458.21105098724365, "seconds_per_optimizer_step": 25.456169499291313} | |
| {"kind": "train", "optimizer_step": 19, "micro_step": 19, "loss": 0.6909667253494263, "mean_cosine": 0.6545166373252869, "fve_rawvar": 0.04479868176015478, "learning_rate": 6.092248796292334e-06, "grad_norm": 2.0003843307495117, "pred_norm_raw": 100.86968994140625, "gold_norm_raw": 111.66014862060547, "backbone_norm_raw": 100.86221313476562, "elapsed_seconds": 483.6096498966217, "seconds_per_optimizer_step": 25.453139468243247} | |
| {"kind": "train", "optimizer_step": 20, "micro_step": 20, "loss": 0.6351891756057739, "mean_cosine": 0.682405412197113, "fve_rawvar": 0.12190628635049428, "learning_rate": 6.085752776663853e-06, "grad_norm": 2.2089600563049316, "pred_norm_raw": 98.62138366699219, "gold_norm_raw": 118.05543518066406, "backbone_norm_raw": 98.61036682128906, "elapsed_seconds": 508.87994742393494, "seconds_per_optimizer_step": 25.443997371196748} | |
| {"kind": "train", "optimizer_step": 21, "micro_step": 21, "loss": 0.6675465703010559, "mean_cosine": 0.666226714849472, "fve_rawvar": 0.07717500634260377, "learning_rate": 6.078578937624665e-06, "grad_norm": 1.9370076656341553, "pred_norm_raw": 103.82323455810547, "gold_norm_raw": 116.48667907714844, "backbone_norm_raw": 103.82438659667969, "elapsed_seconds": 534.1755700111389, "seconds_per_optimizer_step": 25.43693190529233} | |
| {"kind": "train", "optimizer_step": 22, "micro_step": 22, "loss": 0.6450161933898926, "mean_cosine": 0.6774919033050537, "fve_rawvar": 0.10832129014534475, "learning_rate": 6.070729067045856e-06, "grad_norm": 1.9814116954803467, "pred_norm_raw": 107.48855590820312, "gold_norm_raw": 120.78463745117188, "backbone_norm_raw": 107.47308349609375, "elapsed_seconds": 559.6294713020325, "seconds_per_optimizer_step": 25.437703241001476} | |
| {"kind": "train", "optimizer_step": 23, "micro_step": 23, "loss": 0.7521781921386719, "mean_cosine": 0.6239109039306641, "fve_rawvar": -0.03982083988021601, "learning_rate": 6.062205121279746e-06, "grad_norm": 1.9361494779586792, "pred_norm_raw": 102.41910552978516, "gold_norm_raw": 124.81910705566406, "backbone_norm_raw": 102.42771911621094, "elapsed_seconds": 585.04483294487, "seconds_per_optimizer_step": 25.43673186716826} | |
| {"kind": "train", "optimizer_step": 24, "micro_step": 24, "loss": 0.7084640264511108, "mean_cosine": 0.6457679867744446, "fve_rawvar": 0.02061018690967331, "learning_rate": 6.0530092246723204e-06, "grad_norm": 1.9740705490112305, "pred_norm_raw": 102.04869079589844, "gold_norm_raw": 126.1873779296875, "backbone_norm_raw": 102.07855224609375, "elapsed_seconds": 610.5131788253784, "seconds_per_optimizer_step": 25.438049117724102} | |
| {"kind": "train", "optimizer_step": 25, "micro_step": 25, "loss": 0.6475567817687988, "mean_cosine": 0.6762216091156006, "fve_rawvar": 0.10480914798334862, "learning_rate": 6.043143669033803e-06, "grad_norm": 2.8715624809265137, "pred_norm_raw": 100.26058197021484, "gold_norm_raw": 121.51753234863281, "backbone_norm_raw": 100.22938537597656, "elapsed_seconds": 635.683589220047, "seconds_per_optimizer_step": 25.42734356880188} | |
| {"kind": "eval", "optimizer_step": 25, "real_loss": 0.7131928747726811, "shuffled_loss": 1.2179760055409536, "conditioning_gap": 0.5047831307682725, "mean_cosine": 0.6434035626136594, "fve_rawvar": 0.0070482821233267545, "eval_seconds": 35.44353628158569, "elapsed_seconds": 671.1334578990936} | |
| {"kind": "train", "optimizer_step": 26, "micro_step": 26, "loss": 0.7125236988067627, "mean_cosine": 0.6437381505966187, "fve_rawvar": 0.014998043454025, "learning_rate": 6.032610913067489e-06, "grad_norm": 2.517085313796997, "pred_norm_raw": 99.27607727050781, "gold_norm_raw": 119.54208374023438, "backbone_norm_raw": 99.22003173828125, "elapsed_seconds": 696.0057671070099, "seconds_per_optimizer_step": 26.76945258103884} | |
| {"kind": "train", "optimizer_step": 27, "micro_step": 27, "loss": 0.67779541015625, "mean_cosine": 0.661102294921875, "fve_rawvar": 0.06300687786266901, "learning_rate": 6.021413581756979e-06, "grad_norm": 2.1746737957000732, "pred_norm_raw": 100.83565521240234, "gold_norm_raw": 121.84138488769531, "backbone_norm_raw": 100.8027572631836, "elapsed_seconds": 721.3011147975922, "seconds_per_optimizer_step": 26.714856103614526} | |
| {"kind": "train", "optimizer_step": 28, "micro_step": 28, "loss": 0.6540820598602295, "mean_cosine": 0.6729589700698853, "fve_rawvar": 0.09578851933923538, "learning_rate": 6.009554465711984e-06, "grad_norm": 1.9010159969329834, "pred_norm_raw": 98.5272216796875, "gold_norm_raw": 117.05833435058594, "backbone_norm_raw": 98.47200012207031, "elapsed_seconds": 746.5909724235535, "seconds_per_optimizer_step": 26.663963300841196} | |
| {"kind": "train", "optimizer_step": 29, "micro_step": 29, "loss": 0.7444702386856079, "mean_cosine": 0.627764880657196, "fve_rawvar": -0.02916526555342802, "learning_rate": 5.997036520472847e-06, "grad_norm": 2.16192364692688, "pred_norm_raw": 105.82278442382812, "gold_norm_raw": 121.94086456298828, "backbone_norm_raw": 105.7108154296875, "elapsed_seconds": 771.7679967880249, "seconds_per_optimizer_step": 26.612689544414653} | |
| {"kind": "train", "optimizer_step": 30, "micro_step": 30, "loss": 0.5827499628067017, "mean_cosine": 0.7086250185966492, "fve_rawvar": 0.1943989308664842, "learning_rate": 5.983862865773954e-06, "grad_norm": 1.825636625289917, "pred_norm_raw": 106.53228759765625, "gold_norm_raw": 121.80442810058594, "backbone_norm_raw": 106.56011199951172, "elapsed_seconds": 797.0864942073822, "seconds_per_optimizer_step": 26.56954980691274} | |
| {"kind": "train", "optimizer_step": 31, "micro_step": 31, "loss": 0.6520893573760986, "mean_cosine": 0.6739553213119507, "fve_rawvar": 0.09854325696967492, "learning_rate": 5.970036784766239e-06, "grad_norm": 1.6791667938232422, "pred_norm_raw": 104.11644744873047, "gold_norm_raw": 114.83042907714844, "backbone_norm_raw": 104.06278991699219, "elapsed_seconds": 822.3882730007172, "seconds_per_optimizer_step": 26.52865396776507} | |
| {"kind": "train", "optimizer_step": 32, "micro_step": 32, "loss": 0.642500638961792, "mean_cosine": 0.678749680519104, "fve_rawvar": 0.11179882505067695, "learning_rate": 5.955561723198948e-06, "grad_norm": 2.2197816371917725, "pred_norm_raw": 106.567626953125, "gold_norm_raw": 116.8162841796875, "backbone_norm_raw": 106.45417785644531, "elapsed_seconds": 847.8082056045532, "seconds_per_optimizer_step": 26.494006425142288} | |
| {"kind": "train", "optimizer_step": 33, "micro_step": 33, "loss": 0.6290953755378723, "mean_cosine": 0.6854523122310638, "fve_rawvar": 0.13033043420653756, "learning_rate": 5.940441288560889e-06, "grad_norm": 1.8194657564163208, "pred_norm_raw": 106.89155578613281, "gold_norm_raw": 127.32917785644531, "backbone_norm_raw": 106.90568542480469, "elapsed_seconds": 872.961344242096, "seconds_per_optimizer_step": 26.453374067942303} | |
| {"kind": "train", "optimizer_step": 34, "micro_step": 34, "loss": 0.6112866997718811, "mean_cosine": 0.6943566501140594, "fve_rawvar": 0.15494937741133274, "learning_rate": 5.924679249181369e-06, "grad_norm": 2.1081345081329346, "pred_norm_raw": 100.71611785888672, "gold_norm_raw": 119.764404296875, "backbone_norm_raw": 100.5321044921875, "elapsed_seconds": 898.0817677974701, "seconds_per_optimizer_step": 26.414169641102063} | |
| {"kind": "train", "optimizer_step": 35, "micro_step": 35, "loss": 0.5988593101501465, "mean_cosine": 0.7005703449249268, "fve_rawvar": 0.17212916120332067, "learning_rate": 5.908279533291049e-06, "grad_norm": 1.9720251560211182, "pred_norm_raw": 107.78157043457031, "gold_norm_raw": 117.61307525634766, "backbone_norm_raw": 107.56428527832031, "elapsed_seconds": 923.3976368904114, "seconds_per_optimizer_step": 26.382789625440324} | |
| {"kind": "train", "optimizer_step": 36, "micro_step": 36, "loss": 0.5885001420974731, "mean_cosine": 0.7057499289512634, "fve_rawvar": 0.1864498088072667, "learning_rate": 5.89124622804294e-06, "grad_norm": 1.9541938304901123, "pred_norm_raw": 105.17112731933594, "gold_norm_raw": 121.21087646484375, "backbone_norm_raw": 105.16098022460938, "elapsed_seconds": 948.7105586528778, "seconds_per_optimizer_step": 26.35307107369105} | |
| {"kind": "train", "optimizer_step": 37, "micro_step": 37, "loss": 0.6142199039459229, "mean_cosine": 0.6928900480270386, "fve_rawvar": 0.1508944781073258, "learning_rate": 5.873583578493811e-06, "grad_norm": 1.9241260290145874, "pred_norm_raw": 99.96559143066406, "gold_norm_raw": 127.46773529052734, "backbone_norm_raw": 99.96875, "elapsed_seconds": 973.9212021827698, "seconds_per_optimizer_step": 26.322194653588372} | |
| {"kind": "train", "optimizer_step": 38, "micro_step": 38, "loss": 0.6056602001190186, "mean_cosine": 0.6971698999404907, "fve_rawvar": 0.1627275231429849, "learning_rate": 5.85529598654622e-06, "grad_norm": 2.344752073287964, "pred_norm_raw": 96.16534423828125, "gold_norm_raw": 121.88839721679688, "backbone_norm_raw": 96.14920043945312, "elapsed_seconds": 999.0793800354004, "seconds_per_optimizer_step": 26.291562632510537} | |
| {"kind": "train", "optimizer_step": 39, "micro_step": 39, "loss": 0.6898245811462402, "mean_cosine": 0.6550877094268799, "fve_rawvar": 0.046377596646326125, "learning_rate": 5.8363880098514745e-06, "grad_norm": 1.9951419830322266, "pred_norm_raw": 103.49051666259766, "gold_norm_raw": 128.1720733642578, "backbone_norm_raw": 103.47264862060547, "elapsed_seconds": 1024.2226996421814, "seconds_per_optimizer_step": 26.262120503645676} | |
| {"kind": "train", "optimizer_step": 40, "micro_step": 40, "loss": 0.6944954991340637, "mean_cosine": 0.6527522504329681, "fve_rawvar": 0.03992045934046973, "learning_rate": 5.816864360673766e-06, "grad_norm": 2.5259900093078613, "pred_norm_raw": 102.67643737792969, "gold_norm_raw": 130.64260864257812, "backbone_norm_raw": 102.68668365478516, "elapsed_seconds": 1049.7471342086792, "seconds_per_optimizer_step": 26.24367835521698} | |
| {"kind": "train", "optimizer_step": 41, "micro_step": 41, "loss": 0.5674607157707214, "mean_cosine": 0.7162696421146393, "fve_rawvar": 0.21553498328098708, "learning_rate": 5.796729904715778e-06, "grad_norm": 1.7733614444732666, "pred_norm_raw": 102.52848052978516, "gold_norm_raw": 118.12724304199219, "backbone_norm_raw": 102.4255142211914, "elapsed_seconds": 1074.936930179596, "seconds_per_optimizer_step": 26.217973906819413} | |
| {"kind": "train", "optimizer_step": 42, "micro_step": 42, "loss": 0.7577511668205261, "mean_cosine": 0.6211244165897369, "fve_rawvar": -0.04752499200119176, "learning_rate": 5.775989659906047e-06, "grad_norm": 2.0650863647460938, "pred_norm_raw": 102.1120834350586, "gold_norm_raw": 121.06442260742188, "backbone_norm_raw": 102.10812377929688, "elapsed_seconds": 1100.178066253662, "seconds_per_optimizer_step": 26.19471586318243} | |
| {"kind": "train", "optimizer_step": 43, "micro_step": 43, "loss": 0.7769206762313843, "mean_cosine": 0.6115396618843079, "fve_rawvar": -0.07402516919858582, "learning_rate": 5.7546487951483926e-06, "grad_norm": 2.336118459701538, "pred_norm_raw": 100.36190795898438, "gold_norm_raw": 124.70194244384766, "backbone_norm_raw": 100.36006927490234, "elapsed_seconds": 1125.3179605007172, "seconds_per_optimizer_step": 26.170185127923656} | |
| {"kind": "train", "optimizer_step": 44, "micro_step": 44, "loss": 0.5073883533477783, "mean_cosine": 0.7463058233261108, "fve_rawvar": 0.29857979234492427, "learning_rate": 5.7327126290337155e-06, "grad_norm": 1.932560920715332, "pred_norm_raw": 110.7054214477539, "gold_norm_raw": 121.70668029785156, "backbone_norm_raw": 110.57351684570312, "elapsed_seconds": 1150.5500891208649, "seconds_per_optimizer_step": 26.14886566183784} | |
| {"kind": "train", "optimizer_step": 45, "micro_step": 45, "loss": 0.5395911335945129, "mean_cosine": 0.7302044332027435, "fve_rawvar": 0.25406225334604793, "learning_rate": 5.710186628514498e-06, "grad_norm": 1.6946038007736206, "pred_norm_raw": 101.68635559082031, "gold_norm_raw": 113.21736145019531, "backbone_norm_raw": 101.67901611328125, "elapsed_seconds": 1175.8611323833466, "seconds_per_optimizer_step": 26.130247386296592} | |
| {"kind": "train", "optimizer_step": 46, "micro_step": 46, "loss": 0.5966532230377197, "mean_cosine": 0.7016733884811401, "fve_rawvar": 0.17517888449770413, "learning_rate": 5.687076407542319e-06, "grad_norm": 2.121731758117676, "pred_norm_raw": 104.03169250488281, "gold_norm_raw": 116.31659698486328, "backbone_norm_raw": 103.90158081054688, "elapsed_seconds": 1201.130057811737, "seconds_per_optimizer_step": 26.11152299590733} | |
| {"kind": "train", "optimizer_step": 47, "micro_step": 47, "loss": 0.6826276779174805, "mean_cosine": 0.6586861610412598, "fve_rawvar": 0.056326688547791326, "learning_rate": 5.6633877256687445e-06, "grad_norm": 1.8443262577056885, "pred_norm_raw": 108.89620971679688, "gold_norm_raw": 122.39859008789062, "backbone_norm_raw": 108.79562377929688, "elapsed_seconds": 1226.3692569732666, "seconds_per_optimizer_step": 26.092962914324822} | |
| {"kind": "train", "optimizer_step": 48, "micro_step": 48, "loss": 0.6360354423522949, "mean_cosine": 0.6819822788238525, "fve_rawvar": 0.12073639627879706, "learning_rate": 5.639126486609923e-06, "grad_norm": 1.7780452966690063, "pred_norm_raw": 111.28812408447266, "gold_norm_raw": 129.20394897460938, "backbone_norm_raw": 111.24961853027344, "elapsed_seconds": 1251.876780986786, "seconds_per_optimizer_step": 26.08076627055804} | |
| {"kind": "train", "optimizer_step": 49, "micro_step": 49, "loss": 0.5557219982147217, "mean_cosine": 0.7221390008926392, "fve_rawvar": 0.23176273792179958, "learning_rate": 5.614298736775256e-06, "grad_norm": 1.5792129039764404, "pred_norm_raw": 110.5672378540039, "gold_norm_raw": 111.16402435302734, "backbone_norm_raw": 110.14002990722656, "elapsed_seconds": 1276.938907623291, "seconds_per_optimizer_step": 26.059977706597778} | |
| {"kind": "train", "optimizer_step": 50, "micro_step": 50, "loss": 0.6325961351394653, "mean_cosine": 0.6837019324302673, "fve_rawvar": 0.12549093895502395, "learning_rate": 5.588910663760508e-06, "grad_norm": 1.7780529260635376, "pred_norm_raw": 106.066162109375, "gold_norm_raw": 113.57713317871094, "backbone_norm_raw": 105.71068572998047, "elapsed_seconds": 1302.1018648147583, "seconds_per_optimizer_step": 26.042037296295167} | |
| {"kind": "eval", "optimizer_step": 50, "real_loss": 0.6258179141829411, "shuffled_loss": 1.239109353058868, "conditioning_gap": 0.6132914388759269, "mean_cosine": 0.6870910429085295, "fve_rawvar": 0.1286971660169609, "eval_seconds": 35.45733118057251, "elapsed_seconds": 1337.5682740211487} | |
| {"kind": "train", "optimizer_step": 51, "micro_step": 51, "loss": 0.6706683039665222, "mean_cosine": 0.6646658480167389, "fve_rawvar": 0.07285948143662635, "learning_rate": 5.562968594805728e-06, "grad_norm": 2.105349540710449, "pred_norm_raw": 97.36727905273438, "gold_norm_raw": 119.62098693847656, "backbone_norm_raw": 97.38508605957031, "elapsed_seconds": 1362.2932064533234, "seconds_per_optimizer_step": 26.71163149908477} | |
| {"kind": "train", "optimizer_step": 52, "micro_step": 52, "loss": 0.6085346937179565, "mean_cosine": 0.6957326531410217, "fve_rawvar": 0.1587537860956777, "learning_rate": 5.536478995218367e-06, "grad_norm": 1.7129786014556885, "pred_norm_raw": 103.27052307128906, "gold_norm_raw": 114.98359680175781, "backbone_norm_raw": 102.87724304199219, "elapsed_seconds": 1387.6531982421875, "seconds_per_optimizer_step": 26.685638427734375} | |
| {"kind": "train", "optimizer_step": 53, "micro_step": 53, "loss": 0.6226520538330078, "mean_cosine": 0.6886739730834961, "fve_rawvar": 0.13923776528419196, "learning_rate": 5.509448466761989e-06, "grad_norm": 1.8076311349868774, "pred_norm_raw": 101.13676452636719, "gold_norm_raw": 122.95040893554688, "backbone_norm_raw": 101.09259033203125, "elapsed_seconds": 1412.8272886276245, "seconds_per_optimizer_step": 26.657118653351407} | |
| {"kind": "train", "optimizer_step": 54, "micro_step": 54, "loss": 0.7030808925628662, "mean_cosine": 0.6484595537185669, "fve_rawvar": 0.028051900667615204, "learning_rate": 5.481883746010971e-06, "grad_norm": 1.8493863344192505, "pred_norm_raw": 96.95722198486328, "gold_norm_raw": 118.25055694580078, "backbone_norm_raw": 96.92965698242188, "elapsed_seconds": 1438.1249175071716, "seconds_per_optimizer_step": 26.631942916799474} | |
| {"kind": "train", "optimizer_step": 55, "micro_step": 55, "loss": 0.6949566602706909, "mean_cosine": 0.6525216698646545, "fve_rawvar": 0.0392829442337842, "learning_rate": 5.45379170267161e-06, "grad_norm": 1.728350281715393, "pred_norm_raw": 101.213623046875, "gold_norm_raw": 118.63198852539062, "backbone_norm_raw": 101.1964111328125, "elapsed_seconds": 1463.3859331607819, "seconds_per_optimizer_step": 26.60701696655967} | |
| {"kind": "train", "optimizer_step": 56, "micro_step": 56, "loss": 0.7484049797058105, "mean_cosine": 0.6257975101470947, "fve_rawvar": -0.03460470231867796, "learning_rate": 5.425179337870047e-06, "grad_norm": 1.8764654397964478, "pred_norm_raw": 97.10830688476562, "gold_norm_raw": 122.66972351074219, "backbone_norm_raw": 97.02394104003906, "elapsed_seconds": 1488.5520730018616, "seconds_per_optimizer_step": 26.581287017890386} | |
| {"kind": "train", "optimizer_step": 57, "micro_step": 57, "loss": 0.7482532858848572, "mean_cosine": 0.6258733570575714, "fve_rawvar": -0.03439499882294128, "learning_rate": 5.396053782407439e-06, "grad_norm": 1.5717891454696655, "pred_norm_raw": 98.48443603515625, "gold_norm_raw": 123.0322036743164, "backbone_norm_raw": 98.44136047363281, "elapsed_seconds": 1513.7118430137634, "seconds_per_optimizer_step": 26.55634812304848} | |
| {"kind": "train", "optimizer_step": 58, "micro_step": 58, "loss": 0.7465546131134033, "mean_cosine": 0.6267226934432983, "fve_rawvar": -0.032046731661841354, "learning_rate": 5.366422294982817e-06, "grad_norm": 1.804006814956665, "pred_norm_raw": 105.04405212402344, "gold_norm_raw": 114.2374267578125, "backbone_norm_raw": 104.8247299194336, "elapsed_seconds": 1538.9814867973328, "seconds_per_optimizer_step": 26.534163565471253} | |
| {"kind": "train", "optimizer_step": 59, "micro_step": 59, "loss": 0.6430456638336182, "mean_cosine": 0.6784771680831909, "fve_rawvar": 0.11104537563416772, "learning_rate": 5.33629226038406e-06, "grad_norm": 1.7622650861740112, "pred_norm_raw": 98.96311950683594, "gold_norm_raw": 118.75979614257812, "backbone_norm_raw": 98.82351684570312, "elapsed_seconds": 1564.298670053482, "seconds_per_optimizer_step": 26.51353678056749} | |
| {"kind": "train", "optimizer_step": 60, "micro_step": 60, "loss": 0.5692216157913208, "mean_cosine": 0.7153891921043396, "fve_rawvar": 0.21310069236760854, "learning_rate": 5.305671187647459e-06, "grad_norm": 1.9958264827728271, "pred_norm_raw": 102.2381591796875, "gold_norm_raw": 121.15351104736328, "backbone_norm_raw": 102.23800659179688, "elapsed_seconds": 1589.4732682704926, "seconds_per_optimizer_step": 26.491221137841542} | |
| {"kind": "train", "optimizer_step": 61, "micro_step": 61, "loss": 0.6295714378356934, "mean_cosine": 0.6852142810821533, "fve_rawvar": 0.12967231954231406, "learning_rate": 5.274566708186299e-06, "grad_norm": 2.1204912662506104, "pred_norm_raw": 105.11764526367188, "gold_norm_raw": 131.4539794921875, "backbone_norm_raw": 104.92434692382812, "elapsed_seconds": 1614.6448292732239, "seconds_per_optimizer_step": 26.469587365134817} | |
| {"kind": "train", "optimizer_step": 62, "micro_step": 62, "loss": 0.581055760383606, "mean_cosine": 0.709472119808197, "fve_rawvar": 0.19674101816032274, "learning_rate": 5.242986573888959e-06, "grad_norm": 1.7843157052993774, "pred_norm_raw": 102.42176818847656, "gold_norm_raw": 113.32070922851562, "backbone_norm_raw": 102.23580932617188, "elapsed_seconds": 1640.1271317005157, "seconds_per_optimizer_step": 26.45366341452445} | |
| {"kind": "train", "optimizer_step": 63, "micro_step": 63, "loss": 0.6222929954528809, "mean_cosine": 0.6888535022735596, "fve_rawvar": 0.1397341322226272, "learning_rate": 5.210938655186973e-06, "grad_norm": 1.9486767053604126, "pred_norm_raw": 100.26795959472656, "gold_norm_raw": 122.63520812988281, "backbone_norm_raw": 100.25154113769531, "elapsed_seconds": 1665.403963804245, "seconds_per_optimizer_step": 26.434983552448333} | |
| {"kind": "train", "optimizer_step": 64, "micro_step": 64, "loss": 0.5291849970817566, "mean_cosine": 0.7354075014591217, "fve_rawvar": 0.2684478307553536, "learning_rate": 5.178430939093556e-06, "grad_norm": 2.2998945713043213, "pred_norm_raw": 110.26012420654297, "gold_norm_raw": 117.6767578125, "backbone_norm_raw": 109.83662414550781, "elapsed_seconds": 1690.7911858558655, "seconds_per_optimizer_step": 26.418612278997898} | |
| {"kind": "train", "optimizer_step": 65, "micro_step": 65, "loss": 0.5442827939987183, "mean_cosine": 0.7278586030006409, "fve_rawvar": 0.2475764414560987, "learning_rate": 5.145471527213072e-06, "grad_norm": 2.230109691619873, "pred_norm_raw": 108.10166931152344, "gold_norm_raw": 118.75129699707031, "backbone_norm_raw": 107.72930908203125, "elapsed_seconds": 1716.1579852104187, "seconds_per_optimizer_step": 26.40243054169875} | |
| {"kind": "train", "optimizer_step": 66, "micro_step": 66, "loss": 0.725206196308136, "mean_cosine": 0.637396901845932, "fve_rawvar": -0.0025344047630131428, "learning_rate": 5.112068633721945e-06, "grad_norm": 2.1545047760009766, "pred_norm_raw": 108.52919006347656, "gold_norm_raw": 115.00000762939453, "backbone_norm_raw": 108.1173095703125, "elapsed_seconds": 1741.3334619998932, "seconds_per_optimizer_step": 26.383840333331715} | |
| {"kind": "train", "optimizer_step": 67, "micro_step": 67, "loss": 0.576556384563446, "mean_cosine": 0.711721807718277, "fve_rawvar": 0.20296101335980188, "learning_rate": 5.078230583321516e-06, "grad_norm": 1.8297744989395142, "pred_norm_raw": 110.52845001220703, "gold_norm_raw": 121.97720336914062, "backbone_norm_raw": 110.37789916992188, "elapsed_seconds": 1766.5969076156616, "seconds_per_optimizer_step": 26.36711802411435} | |
| {"kind": "train", "optimizer_step": 68, "micro_step": 68, "loss": 0.6361187696456909, "mean_cosine": 0.6819406151771545, "fve_rawvar": 0.12062120355304462, "learning_rate": 5.043965809163347e-06, "grad_norm": 1.9296956062316895, "pred_norm_raw": 106.42454528808594, "gold_norm_raw": 115.36369323730469, "backbone_norm_raw": 106.38053894042969, "elapsed_seconds": 1791.7921633720398, "seconds_per_optimizer_step": 26.349884755471173} | |
| {"kind": "train", "optimizer_step": 69, "micro_step": 69, "loss": 0.6143411993980408, "mean_cosine": 0.6928294003009796, "fve_rawvar": 0.1507267977089667, "learning_rate": 5.009282850747509e-06, "grad_norm": 2.017000198364258, "pred_norm_raw": 105.94630432128906, "gold_norm_raw": 116.95088195800781, "backbone_norm_raw": 105.74708557128906, "elapsed_seconds": 1817.12362241745, "seconds_per_optimizer_step": 26.33512496257174} | |
| {"kind": "train", "optimizer_step": 70, "micro_step": 70, "loss": 0.43588146567344666, "mean_cosine": 0.7820592671632767, "fve_rawvar": 0.3974318366820936, "learning_rate": 4.974190351794358e-06, "grad_norm": 1.7069239616394043, "pred_norm_raw": 117.06438446044922, "gold_norm_raw": 122.26090240478516, "backbone_norm_raw": 116.83784484863281, "elapsed_seconds": 1842.2789952754974, "seconds_per_optimizer_step": 26.318271361078533} | |
| {"kind": "train", "optimizer_step": 71, "micro_step": 71, "loss": 0.5521183013916016, "mean_cosine": 0.7239408493041992, "fve_rawvar": 0.2367445349167856, "learning_rate": 4.9386970580903335e-06, "grad_norm": 1.5055296421051025, "pred_norm_raw": 107.16049194335938, "gold_norm_raw": 122.28962707519531, "backbone_norm_raw": 107.06816101074219, "elapsed_seconds": 1867.433034658432, "seconds_per_optimizer_step": 26.301873727583548} | |
| {"kind": "train", "optimizer_step": 72, "micro_step": 72, "loss": 0.6023872494697571, "mean_cosine": 0.6988063752651215, "fve_rawvar": 0.16725209235885785, "learning_rate": 4.902811815308334e-06, "grad_norm": 1.7968939542770386, "pred_norm_raw": 109.18804931640625, "gold_norm_raw": 127.07501983642578, "backbone_norm_raw": 109.1165771484375, "elapsed_seconds": 1892.864737033844, "seconds_per_optimizer_step": 26.289788014358944} | |
| {"kind": "train", "optimizer_step": 73, "micro_step": 73, "loss": 0.5722048282623291, "mean_cosine": 0.7138975858688354, "fve_rawvar": 0.2089766609485042, "learning_rate": 4.866543566803184e-06, "grad_norm": 1.868726134300232, "pred_norm_raw": 103.390380859375, "gold_norm_raw": 113.08027648925781, "backbone_norm_raw": 103.33743286132812, "elapsed_seconds": 1917.940182685852, "seconds_per_optimizer_step": 26.273153187477426} | |
| {"kind": "train", "optimizer_step": 74, "micro_step": 74, "loss": 0.5428111553192139, "mean_cosine": 0.7285944223403931, "fve_rawvar": 0.24961085375854997, "learning_rate": 4.829901351382763e-06, "grad_norm": 1.9909135103225708, "pred_norm_raw": 108.97711181640625, "gold_norm_raw": 120.14556884765625, "backbone_norm_raw": 108.92449951171875, "elapsed_seconds": 1943.3227655887604, "seconds_per_optimizer_step": 26.261118453902167} | |
| {"kind": "train", "optimizer_step": 75, "micro_step": 75, "loss": 0.4078643321990967, "mean_cosine": 0.7960678339004517, "fve_rawvar": 0.4361630835658955, "learning_rate": 4.792894301055345e-06, "grad_norm": 1.6535227298736572, "pred_norm_raw": 110.75369262695312, "gold_norm_raw": 112.76079559326172, "backbone_norm_raw": 110.29114532470703, "elapsed_seconds": 1968.3703999519348, "seconds_per_optimizer_step": 26.2449386660258} | |
| {"kind": "eval", "optimizer_step": 75, "real_loss": 0.5646483074459765, "shuffled_loss": 1.2704381371537845, "conditioning_gap": 0.705789829707808, "mean_cosine": 0.7176758462770118, "fve_rawvar": 0.2138613175947075, "eval_seconds": 35.44633221626282, "elapsed_seconds": 2003.82954621315} | |
| {"kind": "train", "optimizer_step": 76, "micro_step": 76, "loss": 0.5283337235450745, "mean_cosine": 0.7358331382274628, "fve_rawvar": 0.26962464227838356, "learning_rate": 4.755531638753714e-06, "grad_norm": 1.9679906368255615, "pred_norm_raw": 108.97220611572266, "gold_norm_raw": 131.71046447753906, "backbone_norm_raw": 108.90650939941406, "elapsed_seconds": 2028.7489848136902, "seconds_per_optimizer_step": 26.694065589653817} | |
| {"kind": "train", "optimizer_step": 77, "micro_step": 77, "loss": 0.5297950506210327, "mean_cosine": 0.7351024746894836, "fve_rawvar": 0.26760448486974864, "learning_rate": 4.717822676036613e-06, "grad_norm": 1.5741068124771118, "pred_norm_raw": 106.07276916503906, "gold_norm_raw": 121.48896789550781, "backbone_norm_raw": 105.95991516113281, "elapsed_seconds": 2053.8929557800293, "seconds_per_optimizer_step": 26.67393449064973} | |
| {"kind": "train", "optimizer_step": 78, "micro_step": 78, "loss": 0.7840051054954529, "mean_cosine": 0.6079974472522736, "fve_rawvar": -0.08381877563975437, "learning_rate": 4.6797768107681064e-06, "grad_norm": 2.0415875911712646, "pred_norm_raw": 109.43144989013672, "gold_norm_raw": 134.00714111328125, "backbone_norm_raw": 109.33076477050781, "elapsed_seconds": 2079.176697731018, "seconds_per_optimizer_step": 26.656111509372028} | |
| {"kind": "train", "optimizer_step": 79, "micro_step": 79, "loss": 0.5097624659538269, "mean_cosine": 0.7451187670230865, "fve_rawvar": 0.2952977884397425, "learning_rate": 4.641403524775438e-06, "grad_norm": 2.332796335220337, "pred_norm_raw": 110.35908508300781, "gold_norm_raw": 124.29742431640625, "backbone_norm_raw": 110.2917709350586, "elapsed_seconds": 2104.362267255783, "seconds_per_optimizer_step": 26.63749705387067} | |
| {"kind": "train", "optimizer_step": 80, "micro_step": 80, "loss": 0.4851824939250946, "mean_cosine": 0.7574087530374527, "fve_rawvar": 0.32927745898360294, "learning_rate": 4.602712381485962e-06, "grad_norm": 2.2523763179779053, "pred_norm_raw": 110.86607360839844, "gold_norm_raw": 123.83141326904297, "backbone_norm_raw": 110.58136749267578, "elapsed_seconds": 2129.601671934128, "seconds_per_optimizer_step": 26.6200208991766} | |
| {"kind": "train", "optimizer_step": 81, "micro_step": 81, "loss": 0.5168378353118896, "mean_cosine": 0.7415810823440552, "fve_rawvar": 0.2855167065295571, "learning_rate": 4.563713023543734e-06, "grad_norm": 1.9992812871932983, "pred_norm_raw": 111.77242279052734, "gold_norm_raw": 129.76925659179688, "backbone_norm_raw": 111.66683959960938, "elapsed_seconds": 2154.868625640869, "seconds_per_optimizer_step": 26.603316365936656} | |
| {"kind": "train", "optimizer_step": 82, "micro_step": 82, "loss": 0.5836002826690674, "mean_cosine": 0.7081998586654663, "fve_rawvar": 0.19322343771513661, "learning_rate": 4.5244151704063615e-06, "grad_norm": 2.688023090362549, "pred_norm_raw": 112.71170043945312, "gold_norm_raw": 128.46725463867188, "backbone_norm_raw": 112.62637329101562, "elapsed_seconds": 2180.1304421424866, "seconds_per_optimizer_step": 26.58695661149374} | |
| {"kind": "train", "optimizer_step": 83, "micro_step": 83, "loss": 0.5295063257217407, "mean_cosine": 0.7352468371391296, "fve_rawvar": 0.26800362189660454, "learning_rate": 4.484828615922715e-06, "grad_norm": 2.6010351181030273, "pred_norm_raw": 106.71699523925781, "gold_norm_raw": 113.23847961425781, "backbone_norm_raw": 106.25370788574219, "elapsed_seconds": 2205.3717274665833, "seconds_per_optimizer_step": 26.570743704416667} | |
| {"kind": "train", "optimizer_step": 84, "micro_step": 84, "loss": 0.41819775104522705, "mean_cosine": 0.7909011244773865, "fve_rawvar": 0.42187803199737517, "learning_rate": 4.44496322589209e-06, "grad_norm": 1.7720171213150024, "pred_norm_raw": 113.45095825195312, "gold_norm_raw": 120.16653442382812, "backbone_norm_raw": 112.896728515625, "elapsed_seconds": 2230.571366071701, "seconds_per_optimizer_step": 26.554421024663107} | |
| {"kind": "train", "optimizer_step": 85, "micro_step": 85, "loss": 0.5241831541061401, "mean_cosine": 0.7379084229469299, "fve_rawvar": 0.27536244303501356, "learning_rate": 4.404828935605448e-06, "grad_norm": 1.837619662284851, "pred_norm_raw": 111.11434936523438, "gold_norm_raw": 116.13356018066406, "backbone_norm_raw": 110.44722747802734, "elapsed_seconds": 2255.9184234142303, "seconds_per_optimizer_step": 26.54021674604977} | |
| {"kind": "train", "optimizer_step": 86, "micro_step": 86, "loss": 0.5557224750518799, "mean_cosine": 0.7221387624740601, "fve_rawvar": 0.23176207873595844, "learning_rate": 4.364435747369332e-06, "grad_norm": 1.9627474546432495, "pred_norm_raw": 113.50408172607422, "gold_norm_raw": 118.88368225097656, "backbone_norm_raw": 113.21697235107422, "elapsed_seconds": 2281.2001407146454, "seconds_per_optimizer_step": 26.525583031565645} | |
| {"kind": "train", "optimizer_step": 87, "micro_step": 87, "loss": 0.4685713052749634, "mean_cosine": 0.7657143473625183, "fve_rawvar": 0.35224098054553066, "learning_rate": 4.32379372801308e-06, "grad_norm": 1.7465317249298096, "pred_norm_raw": 111.958984375, "gold_norm_raw": 122.11746215820312, "backbone_norm_raw": 111.87660217285156, "elapsed_seconds": 2306.427094221115, "seconds_per_optimizer_step": 26.510656255415117} | |
| {"kind": "train", "optimizer_step": 88, "micro_step": 88, "loss": 0.6599528193473816, "mean_cosine": 0.6700235903263092, "fve_rawvar": 0.08767270566042062, "learning_rate": 4.282913006379962e-06, "grad_norm": 2.4860637187957764, "pred_norm_raw": 101.60989379882812, "gold_norm_raw": 119.44802856445312, "backbone_norm_raw": 101.50062561035156, "elapsed_seconds": 2331.931897878647, "seconds_per_optimizer_step": 26.49922611225735} | |
| {"kind": "train", "optimizer_step": 89, "micro_step": 89, "loss": 0.6043332815170288, "mean_cosine": 0.6978333592414856, "fve_rawvar": 0.16456187254262733, "learning_rate": 4.2418037708028616e-06, "grad_norm": 1.8371984958648682, "pred_norm_raw": 113.18930053710938, "gold_norm_raw": 122.00946044921875, "backbone_norm_raw": 113.11283111572266, "elapsed_seconds": 2357.0120074748993, "seconds_per_optimizer_step": 26.483280982864038} | |
| {"kind": "train", "optimizer_step": 90, "micro_step": 90, "loss": 0.5075557231903076, "mean_cosine": 0.7462221384048462, "fve_rawvar": 0.2983484181146576, "learning_rate": 4.200476266565127e-06, "grad_norm": 2.1529414653778076, "pred_norm_raw": 112.21126556396484, "gold_norm_raw": 126.328857421875, "backbone_norm_raw": 112.08445739746094, "elapsed_seconds": 2382.472135782242, "seconds_per_optimizer_step": 26.471912619802687} | |
| {"kind": "train", "optimizer_step": 91, "micro_step": 91, "loss": 0.574340283870697, "mean_cosine": 0.7128298580646515, "fve_rawvar": 0.20602457955685083, "learning_rate": 4.158940793347232e-06, "grad_norm": 2.671060562133789, "pred_norm_raw": 111.93220520019531, "gold_norm_raw": 128.64813232421875, "backbone_norm_raw": 111.62452697753906, "elapsed_seconds": 2407.91207075119, "seconds_per_optimizer_step": 26.460572206057034} | |
| {"kind": "train", "optimizer_step": 92, "micro_step": 92, "loss": 0.5799520611763, "mean_cosine": 0.71002396941185, "fve_rawvar": 0.19826678618804616, "learning_rate": 4.117207702659881e-06, "grad_norm": 1.7183375358581543, "pred_norm_raw": 108.46582794189453, "gold_norm_raw": 120.63612365722656, "backbone_norm_raw": 108.37561798095703, "elapsed_seconds": 2433.0805354118347, "seconds_per_optimizer_step": 26.44652755882429} | |
| {"kind": "train", "optimizer_step": 93, "micro_step": 93, "loss": 0.46456658840179443, "mean_cosine": 0.7677167057991028, "fve_rawvar": 0.35777715283297884, "learning_rate": 4.0752873952641945e-06, "grad_norm": 1.736515998840332, "pred_norm_raw": 109.74126434326172, "gold_norm_raw": 125.382568359375, "backbone_norm_raw": 109.6393814086914, "elapsed_seconds": 2458.241845369339, "seconds_per_optimizer_step": 26.432708014724074} | |
| {"kind": "train", "optimizer_step": 94, "micro_step": 94, "loss": 0.3479130268096924, "mean_cosine": 0.8260434865951538, "fve_rawvar": 0.519040541824394, "learning_rate": 4.033190318579619e-06, "grad_norm": 1.9495116472244263, "pred_norm_raw": 113.25847625732422, "gold_norm_raw": 128.60821533203125, "backbone_norm_raw": 112.91455841064453, "elapsed_seconds": 2483.6133286952972, "seconds_per_optimizer_step": 26.421418390375504} | |
| {"kind": "train", "optimizer_step": 95, "micro_step": 95, "loss": 0.5879204869270325, "mean_cosine": 0.7060397565364838, "fve_rawvar": 0.18725113159549422, "learning_rate": 3.990926964080219e-06, "grad_norm": 1.9986908435821533, "pred_norm_raw": 107.85401153564453, "gold_norm_raw": 124.27362060546875, "backbone_norm_raw": 107.653076171875, "elapsed_seconds": 2508.7875847816467, "seconds_per_optimizer_step": 26.408290366122596} | |
| {"kind": "train", "optimizer_step": 96, "micro_step": 96, "loss": 0.5089782476425171, "mean_cosine": 0.7455108761787415, "fve_rawvar": 0.2963819019538515, "learning_rate": 3.948507864679974e-06, "grad_norm": 1.9049088954925537, "pred_norm_raw": 113.33940124511719, "gold_norm_raw": 117.53121948242188, "backbone_norm_raw": 112.87054443359375, "elapsed_seconds": 2533.9781947135925, "seconds_per_optimizer_step": 26.395606194933254} | |
| {"kind": "train", "optimizer_step": 97, "micro_step": 97, "loss": 0.48390936851501465, "mean_cosine": 0.7580453157424927, "fve_rawvar": 0.33103744398053325, "learning_rate": 3.905943592107756e-06, "grad_norm": 1.1977386474609375, "pred_norm_raw": 109.423828125, "gold_norm_raw": 120.52708435058594, "backbone_norm_raw": 108.83134460449219, "elapsed_seconds": 2559.4784128665924, "seconds_per_optimizer_step": 26.386375390377243} | |
| {"kind": "train", "optimizer_step": 98, "micro_step": 98, "loss": 0.4467392861843109, "mean_cosine": 0.7766303569078445, "fve_rawvar": 0.3824218454846965, "learning_rate": 3.863244754272636e-06, "grad_norm": 1.8931565284729004, "pred_norm_raw": 114.24064636230469, "gold_norm_raw": 124.92998504638672, "backbone_norm_raw": 114.13909912109375, "elapsed_seconds": 2584.644395828247, "seconds_per_optimizer_step": 26.373922406410685} | |
| {"kind": "train", "optimizer_step": 99, "micro_step": 99, "loss": 0.506241500377655, "mean_cosine": 0.7468792498111725, "fve_rawvar": 0.3001652166912776, "learning_rate": 3.820421992620173e-06, "grad_norm": 2.0425469875335693, "pred_norm_raw": 109.39949798583984, "gold_norm_raw": 122.32945251464844, "backbone_norm_raw": 109.03163146972656, "elapsed_seconds": 2610.038172721863, "seconds_per_optimizer_step": 26.364021946685483} | |
| {"kind": "train", "optimizer_step": 100, "micro_step": 100, "loss": 0.4461316168308258, "mean_cosine": 0.7769341915845871, "fve_rawvar": 0.38326189544109546, "learning_rate": 3.777485979480335e-06, "grad_norm": 1.7678033113479614, "pred_norm_raw": 109.1651840209961, "gold_norm_raw": 119.0023422241211, "backbone_norm_raw": 108.83531188964844, "elapsed_seconds": 2635.1417429447174, "seconds_per_optimizer_step": 26.351417429447174} | |
| {"kind": "eval", "optimizer_step": 100, "real_loss": 0.5208715121779177, "shuffled_loss": 1.2708840957946248, "conditioning_gap": 0.7500125836167071, "mean_cosine": 0.7395642439110411, "fve_rawvar": 0.27481010943227213, "eval_seconds": 35.382399797439575, "elapsed_seconds": 2670.5335142612457} | |
| {"kind": "checkpoint", "optimizer_step": 100, "path": "/home/sparrow/NLAmodel/experiments/Exp18_affect_ar_sft/runs/ar5000_3gpu/checkpoints/step_000100", "elapsed_seconds": 2670.5439023971558} | |
| {"kind": "train", "optimizer_step": 101, "micro_step": 101, "loss": 0.5189357995986938, "mean_cosine": 0.7405321002006531, "fve_rawvar": 0.28261645362466925, "learning_rate": 3.7344474154077353e-06, "grad_norm": 1.598507046699524, "pred_norm_raw": 109.2854995727539, "gold_norm_raw": 118.76364135742188, "backbone_norm_raw": 109.14175415039062, "elapsed_seconds": 2946.0542125701904, "seconds_per_optimizer_step": 29.168853589803867} | |
| {"kind": "train", "optimizer_step": 102, "micro_step": 102, "loss": 0.656960666179657, "mean_cosine": 0.6715196669101715, "fve_rawvar": 0.0918090968140477, "learning_rate": 3.6913170265148324e-06, "grad_norm": 2.0539848804473877, "pred_norm_raw": 104.90629577636719, "gold_norm_raw": 127.16584014892578, "backbone_norm_raw": 104.80441284179688, "elapsed_seconds": 2971.5976781845093, "seconds_per_optimizer_step": 29.133310570436365} | |
| {"kind": "train", "optimizer_step": 103, "micro_step": 103, "loss": 0.44608476758003235, "mean_cosine": 0.7769576162099838, "fve_rawvar": 0.38332666044999486, "learning_rate": 3.6481055617987525e-06, "grad_norm": 1.6146684885025024, "pred_norm_raw": 112.75947570800781, "gold_norm_raw": 123.37484741210938, "backbone_norm_raw": 112.45996856689453, "elapsed_seconds": 2996.966499567032, "seconds_per_optimizer_step": 29.096762131718755} | |
| {"kind": "train", "optimizer_step": 104, "micro_step": 104, "loss": 0.46020904183387756, "mean_cosine": 0.7698954790830612, "fve_rawvar": 0.3638010814438114, "learning_rate": 3.6048237904624197e-06, "grad_norm": 2.259455919265747, "pred_norm_raw": 115.65425109863281, "gold_norm_raw": 122.62709045410156, "backbone_norm_raw": 115.28305053710938, "elapsed_seconds": 3023.6375575065613, "seconds_per_optimizer_step": 29.073438052947704} | |
| {"kind": "train", "optimizer_step": 105, "micro_step": 105, "loss": 0.5158400535583496, "mean_cosine": 0.7420799732208252, "fve_rawvar": 0.28689605290230047, "learning_rate": 3.561482499230636e-06, "grad_norm": 1.746726155281067, "pred_norm_raw": 115.87013244628906, "gold_norm_raw": 127.74365234375, "backbone_norm_raw": 115.547119140625, "elapsed_seconds": 3049.0230660438538, "seconds_per_optimizer_step": 29.03831491470337} | |
| {"kind": "train", "optimizer_step": 106, "micro_step": 106, "loss": 0.5718916058540344, "mean_cosine": 0.7140541970729828, "fve_rawvar": 0.20940966364795255, "learning_rate": 3.5180924896618152e-06, "grad_norm": 2.049307346343994, "pred_norm_raw": 105.66610717773438, "gold_norm_raw": 123.13360595703125, "backbone_norm_raw": 105.54251861572266, "elapsed_seconds": 3074.3418197631836, "seconds_per_optimizer_step": 29.003224714747017} | |
| {"kind": "train", "optimizer_step": 107, "micro_step": 107, "loss": 0.3850994110107422, "mean_cosine": 0.8074502944946289, "fve_rawvar": 0.46763360440428425, "learning_rate": 3.4746645754560005e-06, "grad_norm": 1.4382985830307007, "pred_norm_raw": 117.60586547851562, "gold_norm_raw": 121.52017974853516, "backbone_norm_raw": 117.23640441894531, "elapsed_seconds": 3099.630630493164, "seconds_per_optimizer_step": 28.968510565356674} | |
| {"kind": "train", "optimizer_step": 108, "micro_step": 108, "loss": 0.5406505465507507, "mean_cosine": 0.7296747267246246, "fve_rawvar": 0.2525977072033274, "learning_rate": 3.4312095797598622e-06, "grad_norm": 1.9981483221054077, "pred_norm_raw": 106.45333099365234, "gold_norm_raw": 120.5808334350586, "backbone_norm_raw": 106.30972290039062, "elapsed_seconds": 3125.0451123714447, "seconds_per_optimizer_step": 28.93560289232819} | |
| {"kind": "train", "optimizer_step": 109, "micro_step": 109, "loss": 0.5630080699920654, "mean_cosine": 0.7184959650039673, "fve_rawvar": 0.22169037826802707, "learning_rate": 3.3877383324693506e-06, "grad_norm": 2.061131238937378, "pred_norm_raw": 114.20625305175781, "gold_norm_raw": 122.77481079101562, "backbone_norm_raw": 113.77447509765625, "elapsed_seconds": 3150.250349521637, "seconds_per_optimizer_step": 28.901379353409514} | |
| {"kind": "train", "optimizer_step": 110, "micro_step": 110, "loss": 0.5613721013069153, "mean_cosine": 0.7193139493465424, "fve_rawvar": 0.22395196249100724, "learning_rate": 3.34426166753065e-06, "grad_norm": 1.7786778211593628, "pred_norm_raw": 113.33716583251953, "gold_norm_raw": 124.57211303710938, "backbone_norm_raw": 113.18638610839844, "elapsed_seconds": 3175.609519958496, "seconds_per_optimizer_step": 28.869177454168145} | |
| {"kind": "train", "optimizer_step": 111, "micro_step": 111, "loss": 0.40294137597084045, "mean_cosine": 0.7985293120145798, "fve_rawvar": 0.44296864178794926, "learning_rate": 3.3007904202401383e-06, "grad_norm": 1.9305654764175415, "pred_norm_raw": 114.7490234375, "gold_norm_raw": 120.17280578613281, "backbone_norm_raw": 114.38140869140625, "elapsed_seconds": 3200.9562299251556, "seconds_per_optimizer_step": 28.83744351283924} | |
| {"kind": "train", "optimizer_step": 112, "micro_step": 112, "loss": 0.532511293888092, "mean_cosine": 0.733744353055954, "fve_rawvar": 0.26384951512349475, "learning_rate": 3.257335424544e-06, "grad_norm": 2.4907445907592773, "pred_norm_raw": 112.79995727539062, "gold_norm_raw": 125.44356536865234, "backbone_norm_raw": 112.44734191894531, "elapsed_seconds": 3226.417329788208, "seconds_per_optimizer_step": 28.807297587394714} | |
| {"kind": "train", "optimizer_step": 113, "micro_step": 113, "loss": 0.5051425099372864, "mean_cosine": 0.7474287450313568, "fve_rawvar": 0.30168447525881903, "learning_rate": 3.213907510338185e-06, "grad_norm": 1.9882402420043945, "pred_norm_raw": 111.02737426757812, "gold_norm_raw": 116.21162414550781, "backbone_norm_raw": 110.9146728515625, "elapsed_seconds": 3251.8322763442993, "seconds_per_optimizer_step": 28.77727678180796} | |
| {"kind": "train", "optimizer_step": 114, "micro_step": 114, "loss": 0.40493205189704895, "mean_cosine": 0.7975339740514755, "fve_rawvar": 0.44021670569733495, "learning_rate": 3.1705175007693638e-06, "grad_norm": 1.7472445964813232, "pred_norm_raw": 117.58341979980469, "gold_norm_raw": 121.38955688476562, "backbone_norm_raw": 117.19293212890625, "elapsed_seconds": 3277.192283630371, "seconds_per_optimizer_step": 28.747300733599747} | |
| {"kind": "train", "optimizer_step": 115, "micro_step": 115, "loss": 0.5783724784851074, "mean_cosine": 0.7108137607574463, "fve_rawvar": 0.20045042168530258, "learning_rate": 3.1271762095375816e-06, "grad_norm": 1.9016059637069702, "pred_norm_raw": 111.05543518066406, "gold_norm_raw": 125.66304016113281, "backbone_norm_raw": 110.92030334472656, "elapsed_seconds": 3302.5335314273834, "seconds_per_optimizer_step": 28.717682881977247} | |
| {"kind": "train", "optimizer_step": 116, "micro_step": 116, "loss": 0.6156659126281738, "mean_cosine": 0.6921670436859131, "fve_rawvar": 0.14889549704383986, "learning_rate": 3.0838944382012476e-06, "grad_norm": 1.7644613981246948, "pred_norm_raw": 109.26632690429688, "gold_norm_raw": 125.41831970214844, "backbone_norm_raw": 108.91908264160156, "elapsed_seconds": 3327.989360332489, "seconds_per_optimizer_step": 28.689563451142146} | |
| {"kind": "train", "optimizer_step": 117, "micro_step": 117, "loss": 0.502129852771759, "mean_cosine": 0.7489350736141205, "fve_rawvar": 0.3058492114036184, "learning_rate": 3.0406829734851673e-06, "grad_norm": 1.9265528917312622, "pred_norm_raw": 108.86029052734375, "gold_norm_raw": 113.3955078125, "backbone_norm_raw": 108.35150146484375, "elapsed_seconds": 3353.194650888443, "seconds_per_optimizer_step": 28.659783340926865} | |
| {"kind": "train", "optimizer_step": 118, "micro_step": 118, "loss": 0.5956016778945923, "mean_cosine": 0.7021991610527039, "fve_rawvar": 0.17663255407404455, "learning_rate": 2.9975525845922648e-06, "grad_norm": 2.6344425678253174, "pred_norm_raw": 118.66816711425781, "gold_norm_raw": 130.5911865234375, "backbone_norm_raw": 118.5337905883789, "elapsed_seconds": 3378.658844232559, "seconds_per_optimizer_step": 28.63270206976745} | |
| {"kind": "train", "optimizer_step": 119, "micro_step": 119, "loss": 0.49968335032463074, "mean_cosine": 0.7501583248376846, "fve_rawvar": 0.30923128795931965, "learning_rate": 2.9545140205196656e-06, "grad_norm": 2.1871838569641113, "pred_norm_raw": 117.0770263671875, "gold_norm_raw": 123.31669616699219, "backbone_norm_raw": 116.733642578125, "elapsed_seconds": 3403.9375727176666, "seconds_per_optimizer_step": 28.604517417795517} | |
| {"kind": "train", "optimizer_step": 120, "micro_step": 120, "loss": 0.5028859376907349, "mean_cosine": 0.7485570311546326, "fve_rawvar": 0.30480398985414114, "learning_rate": 2.911578007379827e-06, "grad_norm": 1.7510663270950317, "pred_norm_raw": 111.37747192382812, "gold_norm_raw": 123.0380630493164, "backbone_norm_raw": 111.02338409423828, "elapsed_seconds": 3429.385910511017, "seconds_per_optimizer_step": 28.57821592092514} | |
| {"kind": "train", "optimizer_step": 121, "micro_step": 121, "loss": 0.4258764386177063, "mean_cosine": 0.7870617806911469, "fve_rawvar": 0.41126291520159175, "learning_rate": 2.868755245727364e-06, "grad_norm": 1.591209053993225, "pred_norm_raw": 115.11434173583984, "gold_norm_raw": 116.4453353881836, "backbone_norm_raw": 114.74395751953125, "elapsed_seconds": 3454.7125041484833, "seconds_per_optimizer_step": 28.551343009491596} | |
| {"kind": "train", "optimizer_step": 122, "micro_step": 122, "loss": 0.37358710169792175, "mean_cosine": 0.8132064491510391, "fve_rawvar": 0.4835483693678647, "learning_rate": 2.826056407892245e-06, "grad_norm": 1.7457000017166138, "pred_norm_raw": 123.10231018066406, "gold_norm_raw": 118.01194763183594, "backbone_norm_raw": 122.29793548583984, "elapsed_seconds": 3480.0708606243134, "seconds_per_optimizer_step": 28.52517098872388} | |
| {"kind": "train", "optimizer_step": 123, "micro_step": 123, "loss": 0.5391120910644531, "mean_cosine": 0.7304439544677734, "fve_rawvar": 0.25472448792177904, "learning_rate": 2.7834921353200265e-06, "grad_norm": 2.0223288536071777, "pred_norm_raw": 113.34379577636719, "gold_norm_raw": 121.03617095947266, "backbone_norm_raw": 113.17547607421875, "elapsed_seconds": 3505.4671297073364, "seconds_per_optimizer_step": 28.499732761848264} | |
| {"kind": "train", "optimizer_step": 124, "micro_step": 124, "loss": 0.5299010276794434, "mean_cosine": 0.7350494861602783, "fve_rawvar": 0.2674579808165385, "learning_rate": 2.7410730359197804e-06, "grad_norm": 2.204270362854004, "pred_norm_raw": 110.63052368164062, "gold_norm_raw": 121.89991760253906, "backbone_norm_raw": 110.4263687133789, "elapsed_seconds": 3530.854218482971, "seconds_per_optimizer_step": 28.47463079421751} | |
| {"kind": "train", "optimizer_step": 125, "micro_step": 125, "loss": 0.47612640261650085, "mean_cosine": 0.7619367986917496, "fve_rawvar": 0.34179671647996734, "learning_rate": 2.6988096814203814e-06, "grad_norm": 1.409972071647644, "pred_norm_raw": 114.43336486816406, "gold_norm_raw": 124.53921508789062, "backbone_norm_raw": 114.29777526855469, "elapsed_seconds": 3556.2924370765686, "seconds_per_optimizer_step": 28.45033949661255} | |
| {"kind": "eval", "optimizer_step": 125, "real_loss": 0.49614588026371265, "shuffled_loss": 1.2810611294375525, "conditioning_gap": 0.7849152491738398, "mean_cosine": 0.7519270598681437, "fve_rawvar": 0.30923468033480894, "eval_seconds": 37.36824655532837, "elapsed_seconds": 3593.669800758362} | |
| {"kind": "train", "optimizer_step": 126, "micro_step": 126, "loss": 0.5580563545227051, "mean_cosine": 0.7209718227386475, "fve_rawvar": 0.22853569363612924, "learning_rate": 2.6567126047358072e-06, "grad_norm": 2.034910202026367, "pred_norm_raw": 114.38211059570312, "gold_norm_raw": 114.81745910644531, "backbone_norm_raw": 113.8765640258789, "elapsed_seconds": 3616.63759970665, "seconds_per_optimizer_step": 28.70347301354484} | |
| {"kind": "train", "optimizer_step": 127, "micro_step": 127, "loss": 0.5085186958312988, "mean_cosine": 0.7457406520843506, "fve_rawvar": 0.297017192308323, "learning_rate": 2.614792297340119e-06, "grad_norm": 1.7399241924285889, "pred_norm_raw": 117.09352111816406, "gold_norm_raw": 127.1281509399414, "backbone_norm_raw": 116.98699188232422, "elapsed_seconds": 3642.151849269867, "seconds_per_optimizer_step": 28.67836101787297} | |
| {"kind": "train", "optimizer_step": 128, "micro_step": 128, "loss": 0.49076423048973083, "mean_cosine": 0.7546178847551346, "fve_rawvar": 0.32156119432279495, "learning_rate": 2.573059206652768e-06, "grad_norm": 1.700340747833252, "pred_norm_raw": 113.32966613769531, "gold_norm_raw": 122.98033142089844, "backbone_norm_raw": 113.04193115234375, "elapsed_seconds": 3667.569917678833, "seconds_per_optimizer_step": 28.652889981865883} | |
| {"kind": "train", "optimizer_step": 129, "micro_step": 129, "loss": 0.45663678646087646, "mean_cosine": 0.7716816067695618, "fve_rawvar": 0.36873941337239224, "learning_rate": 2.531523733434872e-06, "grad_norm": 1.8603160381317139, "pred_norm_raw": 109.03221130371094, "gold_norm_raw": 117.60594177246094, "backbone_norm_raw": 108.903564453125, "elapsed_seconds": 3693.0905299186707, "seconds_per_optimizer_step": 28.628608759059464} | |
| {"kind": "train", "optimizer_step": 130, "micro_step": 130, "loss": 0.5184783339500427, "mean_cosine": 0.7407608330249786, "fve_rawvar": 0.2832488600410854, "learning_rate": 2.490196229197138e-06, "grad_norm": 1.6581908464431763, "pred_norm_raw": 116.87257385253906, "gold_norm_raw": 119.44444274902344, "backbone_norm_raw": 116.7265625, "elapsed_seconds": 3718.713844537735, "seconds_per_optimizer_step": 28.60549111182873} | |
| {"kind": "train", "optimizer_step": 131, "micro_step": 131, "loss": 0.6186565160751343, "mean_cosine": 0.6906717419624329, "fve_rawvar": 0.14476124824419667, "learning_rate": 2.4490869936200384e-06, "grad_norm": 2.029797315597534, "pred_norm_raw": 113.39755249023438, "gold_norm_raw": 124.25447082519531, "backbone_norm_raw": 113.0826644897461, "elapsed_seconds": 3744.0941264629364, "seconds_per_optimizer_step": 28.58087119437356} | |
| {"kind": "train", "optimizer_step": 132, "micro_step": 132, "loss": 0.45540177822113037, "mean_cosine": 0.7722991108894348, "fve_rawvar": 0.3704467047011404, "learning_rate": 2.4082062719869205e-06, "grad_norm": 1.8046908378601074, "pred_norm_raw": 113.05699157714844, "gold_norm_raw": 124.67048645019531, "backbone_norm_raw": 112.91140747070312, "elapsed_seconds": 3769.5922286510468, "seconds_per_optimizer_step": 28.55751688372005} | |
| {"kind": "train", "optimizer_step": 133, "micro_step": 133, "loss": 0.48153698444366455, "mean_cosine": 0.7592315077781677, "fve_rawvar": 0.3343170583370406, "learning_rate": 2.3675642526306685e-06, "grad_norm": 1.7830393314361572, "pred_norm_raw": 111.79226684570312, "gold_norm_raw": 110.01797485351562, "backbone_norm_raw": 111.39080810546875, "elapsed_seconds": 3794.867727994919, "seconds_per_optimizer_step": 28.53284006011217} | |
| {"kind": "train", "optimizer_step": 134, "micro_step": 134, "loss": 0.42717793583869934, "mean_cosine": 0.7864110320806503, "fve_rawvar": 0.40946370864710924, "learning_rate": 2.3271710643945513e-06, "grad_norm": 2.0639352798461914, "pred_norm_raw": 115.4554443359375, "gold_norm_raw": 126.514404296875, "backbone_norm_raw": 115.07679748535156, "elapsed_seconds": 3820.21177983284, "seconds_per_optimizer_step": 28.509043133080894} | |
| {"kind": "train", "optimizer_step": 135, "micro_step": 135, "loss": 0.45259708166122437, "mean_cosine": 0.7737014591693878, "fve_rawvar": 0.3743239534209397, "learning_rate": 2.28703677410791e-06, "grad_norm": 1.532267451286316, "pred_norm_raw": 114.2972412109375, "gold_norm_raw": 120.1422119140625, "backbone_norm_raw": 113.94527435302734, "elapsed_seconds": 3845.5616631507874, "seconds_per_optimizer_step": 28.48564194926509} | |
| {"kind": "train", "optimizer_step": 136, "micro_step": 136, "loss": 0.33205533027648926, "mean_cosine": 0.8339723348617554, "fve_rawvar": 0.540962426159856, "learning_rate": 2.2471713840772856e-06, "grad_norm": 1.8485537767410278, "pred_norm_raw": 120.70011901855469, "gold_norm_raw": 121.22454833984375, "backbone_norm_raw": 120.30589294433594, "elapsed_seconds": 3870.8335676193237, "seconds_per_optimizer_step": 28.462011526612674} | |
| {"kind": "train", "optimizer_step": 137, "micro_step": 137, "loss": 0.44883307814598083, "mean_cosine": 0.7755834609270096, "fve_rawvar": 0.3795273604559193, "learning_rate": 2.207584829593639e-06, "grad_norm": 1.5295729637145996, "pred_norm_raw": 112.67488098144531, "gold_norm_raw": 126.70722961425781, "backbone_norm_raw": 112.284423828125, "elapsed_seconds": 3896.3912665843964, "seconds_per_optimizer_step": 28.440812164849607} | |
| {"kind": "train", "optimizer_step": 138, "micro_step": 138, "loss": 0.47428762912750244, "mean_cosine": 0.7628561854362488, "fve_rawvar": 0.34433866068103947, "learning_rate": 2.1682869764562664e-06, "grad_norm": 2.040905475616455, "pred_norm_raw": 111.74540710449219, "gold_norm_raw": 118.80524444580078, "backbone_norm_raw": 111.34391784667969, "elapsed_seconds": 3921.8871445655823, "seconds_per_optimizer_step": 28.41947206206944} | |
| {"kind": "train", "optimizer_step": 139, "micro_step": 139, "loss": 0.3843257427215576, "mean_cosine": 0.8078371286392212, "fve_rawvar": 0.46870313343165637, "learning_rate": 2.129287618514038e-06, "grad_norm": 1.4855191707611084, "pred_norm_raw": 111.07889556884766, "gold_norm_raw": 121.73635864257812, "backbone_norm_raw": 110.6047134399414, "elapsed_seconds": 3947.354155778885, "seconds_per_optimizer_step": 28.39823133653874} | |
| {"kind": "train", "optimizer_step": 140, "micro_step": 140, "loss": 0.5177595615386963, "mean_cosine": 0.7411202192306519, "fve_rawvar": 0.28424250029848763, "learning_rate": 2.0905964752245624e-06, "grad_norm": 1.43108332157135, "pred_norm_raw": 114.74915313720703, "gold_norm_raw": 122.09115600585938, "backbone_norm_raw": 114.13300323486328, "elapsed_seconds": 3972.7263493537903, "seconds_per_optimizer_step": 28.3766167810985} | |
| {"kind": "train", "optimizer_step": 141, "micro_step": 141, "loss": 0.4700297713279724, "mean_cosine": 0.7649851143360138, "fve_rawvar": 0.350224778251943, "learning_rate": 2.052223189231895e-06, "grad_norm": 1.353294849395752, "pred_norm_raw": 117.8793716430664, "gold_norm_raw": 124.99971771240234, "backbone_norm_raw": 117.43331909179688, "elapsed_seconds": 3998.0538399219513, "seconds_per_optimizer_step": 28.35499177249611} | |
| {"kind": "train", "optimizer_step": 142, "micro_step": 142, "loss": 0.5745348334312439, "mean_cosine": 0.712732583284378, "fve_rawvar": 0.20575563173363487, "learning_rate": 2.0141773239633875e-06, "grad_norm": 2.8511929512023926, "pred_norm_raw": 113.11228942871094, "gold_norm_raw": 123.46652221679688, "backbone_norm_raw": 112.70867156982422, "elapsed_seconds": 4023.3528695106506, "seconds_per_optimizer_step": 28.333470912046835} | |
| {"kind": "train", "optimizer_step": 143, "micro_step": 143, "loss": 0.4494915008544922, "mean_cosine": 0.7752542495727539, "fve_rawvar": 0.3786171484065459, "learning_rate": 1.9764683612462854e-06, "grad_norm": 1.85688316822052, "pred_norm_raw": 115.26737213134766, "gold_norm_raw": 122.90711975097656, "backbone_norm_raw": 114.87071228027344, "elapsed_seconds": 4048.804413795471, "seconds_per_optimizer_step": 28.31331757898931} | |
| {"kind": "train", "optimizer_step": 144, "micro_step": 144, "loss": 0.7223207950592041, "mean_cosine": 0.638839602470398, "fve_rawvar": 0.0014544111604117838, "learning_rate": 1.9391056989446543e-06, "grad_norm": 1.9058855772018433, "pred_norm_raw": 108.44219970703125, "gold_norm_raw": 110.27099609375, "backbone_norm_raw": 108.28590393066406, "elapsed_seconds": 4074.076012611389, "seconds_per_optimizer_step": 28.292194532023537} | |
| {"kind": "train", "optimizer_step": 145, "micro_step": 145, "loss": 0.534915566444397, "mean_cosine": 0.7325422167778015, "fve_rawvar": 0.2605258177138562, "learning_rate": 1.902098648617238e-06, "grad_norm": 1.5992106199264526, "pred_norm_raw": 117.32339477539062, "gold_norm_raw": 116.04643249511719, "backbone_norm_raw": 116.98158264160156, "elapsed_seconds": 4099.457980871201, "seconds_per_optimizer_step": 28.27212400600828} | |
| {"kind": "train", "optimizer_step": 146, "micro_step": 146, "loss": 0.39527371525764465, "mean_cosine": 0.8023631423711777, "fve_rawvar": 0.4535685149111546, "learning_rate": 1.8654564331968168e-06, "grad_norm": 1.7182942628860474, "pred_norm_raw": 117.9959716796875, "gold_norm_raw": 117.01209259033203, "backbone_norm_raw": 117.5947265625, "elapsed_seconds": 4124.922250032425, "seconds_per_optimizer_step": 28.25289212350976} | |
| {"kind": "train", "optimizer_step": 147, "micro_step": 147, "loss": 0.5576497912406921, "mean_cosine": 0.7211751043796539, "fve_rawvar": 0.22909773196399563, "learning_rate": 1.829188184691666e-06, "grad_norm": 1.4516698122024536, "pred_norm_raw": 114.82337188720703, "gold_norm_raw": 127.4678955078125, "backbone_norm_raw": 114.6504898071289, "elapsed_seconds": 4150.318575382233, "seconds_per_optimizer_step": 28.233459696477773} | |
| {"kind": "train", "optimizer_step": 148, "micro_step": 148, "loss": 0.5375216603279114, "mean_cosine": 0.7312391698360443, "fve_rawvar": 0.25692311989692307, "learning_rate": 1.7933029419096668e-06, "grad_norm": 1.9893712997436523, "pred_norm_raw": 116.48869323730469, "gold_norm_raw": 131.96524047851562, "backbone_norm_raw": 116.35977172851562, "elapsed_seconds": 4175.784679174423, "seconds_per_optimizer_step": 28.21476134577313} | |
| {"kind": "train", "optimizer_step": 149, "micro_step": 149, "loss": 0.4718136787414551, "mean_cosine": 0.7640931606292725, "fve_rawvar": 0.34775868162172574, "learning_rate": 1.7578096482056422e-06, "grad_norm": 3.301133632659912, "pred_norm_raw": 116.0512466430664, "gold_norm_raw": 125.48312377929688, "backbone_norm_raw": 115.88944244384766, "elapsed_seconds": 4201.270243406296, "seconds_per_optimizer_step": 28.196444586619435} | |
| {"kind": "train", "optimizer_step": 150, "micro_step": 150, "loss": 0.44754713773727417, "mean_cosine": 0.7762264311313629, "fve_rawvar": 0.3813050610723322, "learning_rate": 1.7227171492524907e-06, "grad_norm": 2.1677300930023193, "pred_norm_raw": 116.72453308105469, "gold_norm_raw": 121.63236999511719, "backbone_norm_raw": 116.1937255859375, "elapsed_seconds": 4226.690768003464, "seconds_per_optimizer_step": 28.177938453356425} | |
| {"kind": "eval", "optimizer_step": 150, "real_loss": 0.48247416213982636, "shuffled_loss": 1.2879411060776975, "conditioning_gap": 0.8054669439378712, "mean_cosine": 0.7587629189300868, "fve_rawvar": 0.32826930123138676, "eval_seconds": 37.47536516189575, "elapsed_seconds": 4264.189164876938} | |
| {"kind": "train", "optimizer_step": 151, "micro_step": 151, "loss": 0.42486175894737244, "mean_cosine": 0.7875691205263138, "fve_rawvar": 0.41266562147258323, "learning_rate": 1.6880341908366537e-06, "grad_norm": 1.5247668027877808, "pred_norm_raw": 119.82252502441406, "gold_norm_raw": 121.20367431640625, "backbone_norm_raw": 119.64596557617188, "elapsed_seconds": 4287.108204603195, "seconds_per_optimizer_step": 28.39144506359732} | |
| {"kind": "train", "optimizer_step": 152, "micro_step": 152, "loss": 0.38374051451683044, "mean_cosine": 0.8081297427415848, "fve_rawvar": 0.46951216045440314, "learning_rate": 1.6537694166784848e-06, "grad_norm": 1.556138277053833, "pred_norm_raw": 122.81472778320312, "gold_norm_raw": 126.25965881347656, "backbone_norm_raw": 122.66934204101562, "elapsed_seconds": 4312.555176019669, "seconds_per_optimizer_step": 28.372073526445188} | |
| {"kind": "train", "optimizer_step": 153, "micro_step": 153, "loss": 0.5682870149612427, "mean_cosine": 0.7158564925193787, "fve_rawvar": 0.21439269661639082, "learning_rate": 1.6199313662780547e-06, "grad_norm": 1.4652904272079468, "pred_norm_raw": 114.73634338378906, "gold_norm_raw": 120.86373901367188, "backbone_norm_raw": 114.28296661376953, "elapsed_seconds": 4337.991242170334, "seconds_per_optimizer_step": 28.352883935753816} | |
| {"kind": "train", "optimizer_step": 154, "micro_step": 154, "loss": 0.46194344758987427, "mean_cosine": 0.7690282762050629, "fve_rawvar": 0.3614034165437354, "learning_rate": 1.586528472786928e-06, "grad_norm": 1.6984426975250244, "pred_norm_raw": 118.70093536376953, "gold_norm_raw": 125.69168853759766, "backbone_norm_raw": 118.32301330566406, "elapsed_seconds": 4363.40677857399, "seconds_per_optimizer_step": 28.333810250480454} | |
| {"kind": "train", "optimizer_step": 155, "micro_step": 155, "loss": 0.5776604413986206, "mean_cosine": 0.7111697793006897, "fve_rawvar": 0.20143475094269758, "learning_rate": 1.5535690609064447e-06, "grad_norm": 1.480932354927063, "pred_norm_raw": 111.50988006591797, "gold_norm_raw": 115.45160675048828, "backbone_norm_raw": 111.07354736328125, "elapsed_seconds": 4388.809265613556, "seconds_per_optimizer_step": 28.314898487829392} | |
| {"kind": "train", "optimizer_step": 156, "micro_step": 156, "loss": 0.5142286419868469, "mean_cosine": 0.7428856790065765, "fve_rawvar": 0.28912368905445807, "learning_rate": 1.5210613448130274e-06, "grad_norm": 2.0951321125030518, "pred_norm_raw": 110.08952331542969, "gold_norm_raw": 117.36582946777344, "backbone_norm_raw": 109.60610961914062, "elapsed_seconds": 4414.282188653946, "seconds_per_optimizer_step": 28.296680696499653} | |
| {"kind": "train", "optimizer_step": 157, "micro_step": 157, "loss": 0.6100564002990723, "mean_cosine": 0.6949717998504639, "fve_rawvar": 0.15665015927989878, "learning_rate": 1.489013426111041e-06, "grad_norm": 2.6469240188598633, "pred_norm_raw": 110.72383117675781, "gold_norm_raw": 123.10128021240234, "backbone_norm_raw": 110.5482177734375, "elapsed_seconds": 4439.863956928253, "seconds_per_optimizer_step": 28.27938826068951} | |
| {"kind": "train", "optimizer_step": 158, "micro_step": 158, "loss": 0.45425936579704285, "mean_cosine": 0.7728703171014786, "fve_rawvar": 0.37202599037934747, "learning_rate": 1.4574332918137003e-06, "grad_norm": 1.9359647035598755, "pred_norm_raw": 114.29443359375, "gold_norm_raw": 109.89561462402344, "backbone_norm_raw": 113.6219711303711, "elapsed_seconds": 4465.169616937637, "seconds_per_optimizer_step": 28.26056719580783} | |
| {"kind": "train", "optimizer_step": 159, "micro_step": 159, "loss": 0.4252629280090332, "mean_cosine": 0.7873685359954834, "fve_rawvar": 0.4121110401845457, "learning_rate": 1.4263288123525405e-06, "grad_norm": 2.232926607131958, "pred_norm_raw": 113.86736297607422, "gold_norm_raw": 120.47930145263672, "backbone_norm_raw": 113.50517272949219, "elapsed_seconds": 4490.536180973053, "seconds_per_optimizer_step": 28.242365918069517} | |
| {"kind": "train", "optimizer_step": 160, "micro_step": 160, "loss": 0.42758965492248535, "mean_cosine": 0.7862051725387573, "fve_rawvar": 0.40889454287233473, "learning_rate": 1.3957077396159398e-06, "grad_norm": 1.9220153093338013, "pred_norm_raw": 109.28187561035156, "gold_norm_raw": 122.62353515625, "backbone_norm_raw": 108.90708923339844, "elapsed_seconds": 4515.8059141635895, "seconds_per_optimizer_step": 28.223786963522436} | |
| {"kind": "train", "optimizer_step": 161, "micro_step": 161, "loss": 0.5314302444458008, "mean_cosine": 0.7342848777770996, "fve_rawvar": 0.26534397182376035, "learning_rate": 1.365577705017184e-06, "grad_norm": 1.817719578742981, "pred_norm_raw": 110.00251770019531, "gold_norm_raw": 111.29356384277344, "backbone_norm_raw": 109.5498275756836, "elapsed_seconds": 4541.450423002243, "seconds_per_optimizer_step": 28.207766602498403} | |
| {"kind": "train", "optimizer_step": 162, "micro_step": 162, "loss": 0.6744100451469421, "mean_cosine": 0.6627949774265289, "fve_rawvar": 0.06768685014060849, "learning_rate": 1.3359462175925612e-06, "grad_norm": 2.292527198791504, "pred_norm_raw": 111.92042541503906, "gold_norm_raw": 115.71568298339844, "backbone_norm_raw": 111.50509643554688, "elapsed_seconds": 4566.740352869034, "seconds_per_optimizer_step": 28.189755264623667} | |
| {"kind": "train", "optimizer_step": 163, "micro_step": 163, "loss": 0.420473575592041, "mean_cosine": 0.7897632122039795, "fve_rawvar": 0.41873190277371397, "learning_rate": 1.3068206621299532e-06, "grad_norm": 1.6717045307159424, "pred_norm_raw": 118.28096008300781, "gold_norm_raw": 125.62933349609375, "backbone_norm_raw": 118.10791778564453, "elapsed_seconds": 4592.108218193054, "seconds_per_optimizer_step": 28.172443056399107} | |
| {"kind": "train", "optimizer_step": 164, "micro_step": 164, "loss": 0.5612047910690308, "mean_cosine": 0.7193976044654846, "fve_rawvar": 0.22418325432304376, "learning_rate": 1.2782082973283905e-06, "grad_norm": 1.8289700746536255, "pred_norm_raw": 112.01935577392578, "gold_norm_raw": 113.04441833496094, "backbone_norm_raw": 111.59198760986328, "elapsed_seconds": 4617.498178243637, "seconds_per_optimizer_step": 28.155476696607543} | |
| {"kind": "train", "optimizer_step": 165, "micro_step": 165, "loss": 0.5307600498199463, "mean_cosine": 0.7346199750900269, "fve_rawvar": 0.2662704575235888, "learning_rate": 1.2501162539890288e-06, "grad_norm": 1.8566343784332275, "pred_norm_raw": 113.77515411376953, "gold_norm_raw": 121.44596099853516, "backbone_norm_raw": 113.38858032226562, "elapsed_seconds": 4642.884285211563, "seconds_per_optimizer_step": 28.138692637645836} | |
| {"kind": "train", "optimizer_step": 166, "micro_step": 166, "loss": 0.5371524691581726, "mean_cosine": 0.7314237654209137, "fve_rawvar": 0.25743349453448416, "learning_rate": 1.2225515332380112e-06, "grad_norm": 1.6686100959777832, "pred_norm_raw": 116.92364501953125, "gold_norm_raw": 121.33171081542969, "backbone_norm_raw": 116.70204162597656, "elapsed_seconds": 4668.215626955032, "seconds_per_optimizer_step": 28.121780885271278} | |
| {"kind": "train", "optimizer_step": 167, "micro_step": 167, "loss": 0.6292130351066589, "mean_cosine": 0.6853934824466705, "fve_rawvar": 0.13016778010021768, "learning_rate": 1.1955210047816325e-06, "grad_norm": 1.9191583395004272, "pred_norm_raw": 115.49493408203125, "gold_norm_raw": 120.33241271972656, "backbone_norm_raw": 115.32366943359375, "elapsed_seconds": 4693.552230834961, "seconds_per_optimizer_step": 28.10510317865246} | |
| {"kind": "train", "optimizer_step": 168, "micro_step": 168, "loss": 0.6798319816589355, "mean_cosine": 0.6600840091705322, "fve_rawvar": 0.06019149513483768, "learning_rate": 1.169031405194272e-06, "grad_norm": 1.9186477661132812, "pred_norm_raw": 113.45669555664062, "gold_norm_raw": 123.78755187988281, "backbone_norm_raw": 113.30984497070312, "elapsed_seconds": 4718.953001022339, "seconds_per_optimizer_step": 28.089005958466302} | |
| {"kind": "train", "optimizer_step": 169, "micro_step": 169, "loss": 0.5153127908706665, "mean_cosine": 0.7423436045646667, "fve_rawvar": 0.2876249476462245, "learning_rate": 1.1430893362394926e-06, "grad_norm": 2.0404655933380127, "pred_norm_raw": 109.64677429199219, "gold_norm_raw": 122.03227233886719, "backbone_norm_raw": 109.48683166503906, "elapsed_seconds": 4744.343268632889, "seconds_per_optimizer_step": 28.0730370925023} | |
| {"kind": "train", "optimizer_step": 170, "micro_step": 170, "loss": 0.394478440284729, "mean_cosine": 0.8027607798576355, "fve_rawvar": 0.45466791329695666, "learning_rate": 1.117701263224745e-06, "grad_norm": 1.8281002044677734, "pred_norm_raw": 113.28242492675781, "gold_norm_raw": 119.13021850585938, "backbone_norm_raw": 112.62519836425781, "elapsed_seconds": 4769.742715835571, "seconds_per_optimizer_step": 28.05731009315042} | |
| {"kind": "train", "optimizer_step": 171, "micro_step": 171, "loss": 0.6260144710540771, "mean_cosine": 0.6869927644729614, "fve_rawvar": 0.13458951632486094, "learning_rate": 1.0928735133900783e-06, "grad_norm": 1.6568710803985596, "pred_norm_raw": 109.24223327636719, "gold_norm_raw": 121.96660614013672, "backbone_norm_raw": 108.88497924804688, "elapsed_seconds": 4795.074436426163, "seconds_per_optimizer_step": 28.041370973252413} | |
| {"kind": "train", "optimizer_step": 172, "micro_step": 172, "loss": 0.42725875973701477, "mean_cosine": 0.7863706201314926, "fve_rawvar": 0.40935197664702316, "learning_rate": 1.0686122743312554e-06, "grad_norm": 1.6959376335144043, "pred_norm_raw": 112.50411987304688, "gold_norm_raw": 122.14735412597656, "backbone_norm_raw": 112.26077270507812, "elapsed_seconds": 4820.751872062683, "seconds_per_optimizer_step": 28.027627163155135} | |
| {"kind": "train", "optimizer_step": 173, "micro_step": 173, "loss": 0.49558714032173157, "mean_cosine": 0.7522064298391342, "fve_rawvar": 0.3148939415300511, "learning_rate": 1.0449235924576805e-06, "grad_norm": 1.8279955387115479, "pred_norm_raw": 116.96218872070312, "gold_norm_raw": 121.77667236328125, "backbone_norm_raw": 116.730224609375, "elapsed_seconds": 4846.067575454712, "seconds_per_optimizer_step": 28.011951303206427} | |
| {"kind": "train", "optimizer_step": 174, "micro_step": 174, "loss": 0.47162342071533203, "mean_cosine": 0.764188289642334, "fve_rawvar": 0.3480216967723707, "learning_rate": 1.021813371485502e-06, "grad_norm": 1.8213132619857788, "pred_norm_raw": 114.2742691040039, "gold_norm_raw": 118.76705932617188, "backbone_norm_raw": 113.63863372802734, "elapsed_seconds": 4871.663131475449, "seconds_per_optimizer_step": 27.998063973996832} | |
| {"kind": "train", "optimizer_step": 175, "micro_step": 175, "loss": 0.5382561087608337, "mean_cosine": 0.7308719456195831, "fve_rawvar": 0.255907808904991, "learning_rate": 9.992873709662844e-07, "grad_norm": 2.454197406768799, "pred_norm_raw": 116.00094604492188, "gold_norm_raw": 124.62702941894531, "backbone_norm_raw": 115.61546325683594, "elapsed_seconds": 4896.989147901535, "seconds_per_optimizer_step": 27.982795130865913} | |
| {"kind": "eval", "optimizer_step": 175, "real_loss": 0.4776784409251478, "shuffled_loss": 1.2799315427740414, "conditioning_gap": 0.8022531018488936, "mean_cosine": 0.7611607795374261, "fve_rawvar": 0.3349462041941239, "eval_seconds": 37.54615116119385, "elapsed_seconds": 4934.543691635132} | |
| {"kind": "train", "optimizer_step": 176, "micro_step": 176, "loss": 0.46632295846939087, "mean_cosine": 0.7668385207653046, "fve_rawvar": 0.3553491241850918, "learning_rate": 9.773512048516082e-07, "grad_norm": 1.6588326692581177, "pred_norm_raw": 112.21820068359375, "gold_norm_raw": 122.09342956542969, "backbone_norm_raw": 112.04624938964844, "elapsed_seconds": 4957.487441539764, "seconds_per_optimizer_step": 28.167542281475935} | |
| {"kind": "train", "optimizer_step": 177, "micro_step": 177, "loss": 0.4688485264778137, "mean_cosine": 0.7655757367610931, "fve_rawvar": 0.351857746377094, "learning_rate": 9.560103400939532e-07, "grad_norm": 1.6105231046676636, "pred_norm_raw": 116.63758850097656, "gold_norm_raw": 127.44219970703125, "backbone_norm_raw": 116.45720672607422, "elapsed_seconds": 4982.835569620132, "seconds_per_optimizer_step": 28.151613387684364} | |
| {"kind": "train", "optimizer_step": 178, "micro_step": 178, "loss": 0.5013981461524963, "mean_cosine": 0.7493009269237518, "fve_rawvar": 0.3068607320769636, "learning_rate": 9.352700952842217e-07, "grad_norm": 1.9460939168930054, "pred_norm_raw": 118.34986877441406, "gold_norm_raw": 124.34282684326172, "backbone_norm_raw": 118.1881103515625, "elapsed_seconds": 5008.399860858917, "seconds_per_optimizer_step": 28.137077869993917} | |
| {"kind": "train", "optimizer_step": 179, "micro_step": 179, "loss": 0.3616224229335785, "mean_cosine": 0.8191887885332108, "fve_rawvar": 0.5000884956991831, "learning_rate": 9.151356393262338e-07, "grad_norm": 1.512898325920105, "pred_norm_raw": 113.81578063964844, "gold_norm_raw": 117.9168472290039, "backbone_norm_raw": 113.13893127441406, "elapsed_seconds": 5033.743090391159, "seconds_per_optimizer_step": 28.121469778721558} | |
| {"kind": "train", "optimizer_step": 180, "micro_step": 180, "loss": 0.5236002206802368, "mean_cosine": 0.7381998896598816, "fve_rawvar": 0.27616829772589946, "learning_rate": 8.956119901485255e-07, "grad_norm": 1.9374598264694214, "pred_norm_raw": 112.51471710205078, "gold_norm_raw": 118.78627014160156, "backbone_norm_raw": 112.04627990722656, "elapsed_seconds": 5059.1854610443115, "seconds_per_optimizer_step": 28.10658589469062} | |
| {"kind": "train", "optimizer_step": 181, "micro_step": 181, "loss": 0.6026135087013245, "mean_cosine": 0.6986932456493378, "fve_rawvar": 0.16693930867720108, "learning_rate": 8.767040134537803e-07, "grad_norm": 1.7645376920700073, "pred_norm_raw": 115.74701690673828, "gold_norm_raw": 125.76702880859375, "backbone_norm_raw": 115.58534240722656, "elapsed_seconds": 5084.498535871506, "seconds_per_optimizer_step": 28.091152131886773} | |
| {"kind": "train", "optimizer_step": 182, "micro_step": 182, "loss": 0.39645975828170776, "mean_cosine": 0.8017701208591461, "fve_rawvar": 0.4519289137284762, "learning_rate": 8.584164215061891e-07, "grad_norm": 2.129957437515259, "pred_norm_raw": 113.21420288085938, "gold_norm_raw": 115.73188781738281, "backbone_norm_raw": 112.62677764892578, "elapsed_seconds": 5110.149393081665, "seconds_per_optimizer_step": 28.07774391803113} | |
| {"kind": "train", "optimizer_step": 183, "micro_step": 183, "loss": 0.4793981909751892, "mean_cosine": 0.7603009045124054, "fve_rawvar": 0.3372737540295825, "learning_rate": 8.407537719570597e-07, "grad_norm": 1.6320350170135498, "pred_norm_raw": 113.55995178222656, "gold_norm_raw": 123.31104278564453, "backbone_norm_raw": 113.39146423339844, "elapsed_seconds": 5135.548535346985, "seconds_per_optimizer_step": 28.063106750529972} | |
| {"kind": "train", "optimizer_step": 184, "micro_step": 184, "loss": 0.4247499704360962, "mean_cosine": 0.7876250147819519, "fve_rawvar": 0.41282015935323324, "learning_rate": 8.237204667089518e-07, "grad_norm": 1.9806172847747803, "pred_norm_raw": 122.28204345703125, "gold_norm_raw": 126.07270812988281, "backbone_norm_raw": 121.80586242675781, "elapsed_seconds": 5160.946298122406, "seconds_per_optimizer_step": 28.04862118544786} | |
| {"kind": "train", "optimizer_step": 185, "micro_step": 185, "loss": 0.575512170791626, "mean_cosine": 0.712243914604187, "fve_rawvar": 0.20440454795383356, "learning_rate": 8.0732075081863e-07, "grad_norm": 2.920722246170044, "pred_norm_raw": 108.95732879638672, "gold_norm_raw": 114.37135314941406, "backbone_norm_raw": 108.64566040039062, "elapsed_seconds": 5186.225680112839, "seconds_per_optimizer_step": 28.033652324934263} | |
| {"kind": "train", "optimizer_step": 186, "micro_step": 186, "loss": 0.41684800386428833, "mean_cosine": 0.7915759980678558, "fve_rawvar": 0.4237439399191657, "learning_rate": 7.915587114391101e-07, "grad_norm": 1.4521527290344238, "pred_norm_raw": 113.50039672851562, "gold_norm_raw": 114.69746398925781, "backbone_norm_raw": 112.87659454345703, "elapsed_seconds": 5211.781099081039, "seconds_per_optimizer_step": 28.020328489683006} | |
| {"kind": "train", "optimizer_step": 187, "micro_step": 187, "loss": 0.5338324904441833, "mean_cosine": 0.7330837547779083, "fve_rawvar": 0.2620230759539469, "learning_rate": 7.764382768010518e-07, "grad_norm": 2.065734624862671, "pred_norm_raw": 120.24465942382812, "gold_norm_raw": 126.72710418701172, "backbone_norm_raw": 119.82192993164062, "elapsed_seconds": 5236.99850153923, "seconds_per_optimizer_step": 28.00533958042369} | |
| {"kind": "train", "optimizer_step": 188, "micro_step": 188, "loss": 0.5631997585296631, "mean_cosine": 0.7184001207351685, "fve_rawvar": 0.22142538555985847, "learning_rate": 7.619632152337611e-07, "grad_norm": 1.8011956214904785, "pred_norm_raw": 119.2828369140625, "gold_norm_raw": 119.85792541503906, "backbone_norm_raw": 118.58660888671875, "elapsed_seconds": 5262.340129852295, "seconds_per_optimizer_step": 27.991170903469655} | |
| {"kind": "train", "optimizer_step": 189, "micro_step": 189, "loss": 0.5170170068740845, "mean_cosine": 0.7414914965629578, "fve_rawvar": 0.28526901744972044, "learning_rate": 7.481371342260462e-07, "grad_norm": 1.5033878087997437, "pred_norm_raw": 111.95759582519531, "gold_norm_raw": 116.24213409423828, "backbone_norm_raw": 111.26861572265625, "elapsed_seconds": 5287.581449985504, "seconds_per_optimizer_step": 27.97662142849473} | |
| {"kind": "train", "optimizer_step": 190, "micro_step": 190, "loss": 0.39391031861305237, "mean_cosine": 0.8030448406934738, "fve_rawvar": 0.45545329202764984, "learning_rate": 7.349634795271535e-07, "grad_norm": 2.080827474594116, "pred_norm_raw": 115.29196166992188, "gold_norm_raw": 122.74758911132812, "backbone_norm_raw": 115.1142578125, "elapsed_seconds": 5312.978806257248, "seconds_per_optimizer_step": 27.963046348722358} | |
| {"kind": "train", "optimizer_step": 191, "micro_step": 191, "loss": 0.43367260694503784, "mean_cosine": 0.7831636965274811, "fve_rawvar": 0.40048539149417905, "learning_rate": 7.224455342880155e-07, "grad_norm": 1.5391820669174194, "pred_norm_raw": 119.39823913574219, "gold_norm_raw": 127.77729797363281, "backbone_norm_raw": 119.22830200195312, "elapsed_seconds": 5338.344149589539, "seconds_per_optimizer_step": 27.949445809369312} | |
| {"kind": "train", "optimizer_step": 192, "micro_step": 192, "loss": 0.6891505718231201, "mean_cosine": 0.6554247140884399, "fve_rawvar": 0.0473093558328842, "learning_rate": 7.105864182430214e-07, "grad_norm": 1.7413372993469238, "pred_norm_raw": 110.77816772460938, "gold_norm_raw": 124.00373840332031, "backbone_norm_raw": 110.6325454711914, "elapsed_seconds": 5363.73518037796, "seconds_per_optimizer_step": 27.936120731135208} | |
| {"kind": "train", "optimizer_step": 193, "micro_step": 193, "loss": 0.3596736490726471, "mean_cosine": 0.8201631754636765, "fve_rawvar": 0.5027825058340007, "learning_rate": 6.993890869325113e-07, "grad_norm": 1.6613116264343262, "pred_norm_raw": 114.65162658691406, "gold_norm_raw": 119.65164947509766, "backbone_norm_raw": 114.09098815917969, "elapsed_seconds": 5389.081754207611, "seconds_per_optimizer_step": 27.92270338967674} | |
| {"kind": "train", "optimizer_step": 194, "micro_step": 194, "loss": 0.4496576488018036, "mean_cosine": 0.7751711755990982, "fve_rawvar": 0.3783874633399974, "learning_rate": 6.888563309661964e-07, "grad_norm": 1.9127146005630493, "pred_norm_raw": 118.68946838378906, "gold_norm_raw": 126.97862243652344, "backbone_norm_raw": 118.505859375, "elapsed_seconds": 5414.456577539444, "seconds_per_optimizer_step": 27.909569987316722} | |
| {"kind": "train", "optimizer_step": 195, "micro_step": 195, "loss": 0.5089956521987915, "mean_cosine": 0.7455021739006042, "fve_rawvar": 0.29635784167064716, "learning_rate": 6.789907753276795e-07, "grad_norm": 2.058220386505127, "pred_norm_raw": 113.27311706542969, "gold_norm_raw": 121.06358337402344, "backbone_norm_raw": 113.11637878417969, "elapsed_seconds": 5439.844558954239, "seconds_per_optimizer_step": 27.89663876386789} | |
| {"kind": "train", "optimizer_step": 196, "micro_step": 196, "loss": 0.5500139594078064, "mean_cosine": 0.7249930202960968, "fve_rawvar": 0.23965360443229966, "learning_rate": 6.697948787202542e-07, "grad_norm": 2.1808087825775146, "pred_norm_raw": 115.89771270751953, "gold_norm_raw": 119.5184097290039, "backbone_norm_raw": 115.46401977539062, "elapsed_seconds": 5465.0991587638855, "seconds_per_optimizer_step": 27.88315897328513} | |
| {"kind": "train", "optimizer_step": 197, "micro_step": 197, "loss": 0.4581480920314789, "mean_cosine": 0.7709259539842606, "fve_rawvar": 0.3666501650477749, "learning_rate": 6.61270932954144e-07, "grad_norm": 1.7103626728057861, "pred_norm_raw": 114.61964416503906, "gold_norm_raw": 114.41754150390625, "backbone_norm_raw": 114.23128509521484, "elapsed_seconds": 5490.5435791015625, "seconds_per_optimizer_step": 27.870779589348032} | |
| {"kind": "train", "optimizer_step": 198, "micro_step": 198, "loss": 0.5990786552429199, "mean_cosine": 0.70046067237854, "fve_rawvar": 0.1718259357163615, "learning_rate": 6.53421062375336e-07, "grad_norm": 1.8985682725906372, "pred_norm_raw": 113.58541870117188, "gold_norm_raw": 117.38575744628906, "backbone_norm_raw": 112.99937438964844, "elapsed_seconds": 5515.973284482956, "seconds_per_optimizer_step": 27.8584509317321} | |
| {"kind": "train", "optimizer_step": 199, "micro_step": 199, "loss": 0.5707730650901794, "mean_cosine": 0.7146134674549103, "fve_rawvar": 0.2109559488349838, "learning_rate": 6.462472233361476e-07, "grad_norm": 2.6771271228790283, "pred_norm_raw": 116.88192749023438, "gold_norm_raw": 120.46424865722656, "backbone_norm_raw": 116.69644165039062, "elapsed_seconds": 5541.430006504059, "seconds_per_optimizer_step": 27.846381942231453} | |
| {"kind": "train", "optimizer_step": 200, "micro_step": 200, "loss": 0.5677552223205566, "mean_cosine": 0.7161223888397217, "fve_rawvar": 0.21512785362580644, "learning_rate": 6.397512037076657e-07, "grad_norm": 2.096141815185547, "pred_norm_raw": 106.8187026977539, "gold_norm_raw": 113.22012329101562, "backbone_norm_raw": 106.4495849609375, "elapsed_seconds": 5566.681344985962, "seconds_per_optimizer_step": 27.83340672492981} | |
| {"kind": "eval", "optimizer_step": 200, "real_loss": 0.4746351122028298, "shuffled_loss": 1.2890390286015139, "conditioning_gap": 0.814403916398684, "mean_cosine": 0.7626824438985851, "fve_rawvar": 0.33918331674779634, "eval_seconds": 37.57524752616882, "elapsed_seconds": 5604.265249967575} | |
| {"kind": "checkpoint", "optimizer_step": 200, "path": "/home/sparrow/NLAmodel/experiments/Exp18_affect_ar_sft/runs/ar5000_3gpu/checkpoints/step_000200", "elapsed_seconds": 5604.273499727249} | |
| {"kind": "train", "optimizer_step": 201, "micro_step": 201, "loss": 0.5928037166595459, "mean_cosine": 0.703598141670227, "fve_rawvar": 0.18050049179383665, "learning_rate": 6.339346224341678e-07, "grad_norm": 2.0073347091674805, "pred_norm_raw": 117.34941101074219, "gold_norm_raw": 124.69386291503906, "backbone_norm_raw": 117.15453338623047, "elapsed_seconds": 5873.332209825516, "seconds_per_optimizer_step": 29.220558257838388} | |
| {"kind": "train", "optimizer_step": 202, "micro_step": 202, "loss": 0.42740368843078613, "mean_cosine": 0.7862981557846069, "fve_rawvar": 0.4091516253504087, "learning_rate": 6.28798929129651e-07, "grad_norm": 1.6409484148025513, "pred_norm_raw": 118.95191955566406, "gold_norm_raw": 121.04588317871094, "backbone_norm_raw": 118.33799743652344, "elapsed_seconds": 5898.438188314438, "seconds_per_optimizer_step": 29.200189051061574} | |
| {"kind": "train", "optimizer_step": 203, "micro_step": 203, "loss": 0.48792558908462524, "mean_cosine": 0.7560372054576874, "fve_rawvar": 0.3254853688346657, "learning_rate": 6.243454037165546e-07, "grad_norm": 3.205850839614868, "pred_norm_raw": 121.15409851074219, "gold_norm_raw": 123.73622131347656, "backbone_norm_raw": 120.71719360351562, "elapsed_seconds": 5923.810393810272, "seconds_per_optimizer_step": 29.181331989213163} | |
| {"kind": "train", "optimizer_step": 204, "micro_step": 204, "loss": 0.38240116834640503, "mean_cosine": 0.8087994158267975, "fve_rawvar": 0.4713636898850322, "learning_rate": 6.205751561067775e-07, "grad_norm": 2.0103986263275146, "pred_norm_raw": 122.87776947021484, "gold_norm_raw": 127.00251007080078, "backbone_norm_raw": 122.67658233642578, "elapsed_seconds": 5949.302136659622, "seconds_per_optimizer_step": 29.163245767939326} | |
| {"kind": "train", "optimizer_step": 205, "micro_step": 205, "loss": 0.4771572947502136, "mean_cosine": 0.7614213526248932, "fve_rawvar": 0.34037159789037463, "learning_rate": 6.174891259250671e-07, "grad_norm": 2.4741575717926025, "pred_norm_raw": 119.78939819335938, "gold_norm_raw": 127.46685028076172, "backbone_norm_raw": 119.6037368774414, "elapsed_seconds": 5974.862098932266, "seconds_per_optimizer_step": 29.145668775279347} | |
| {"kind": "train", "optimizer_step": 206, "micro_step": 206, "loss": 0.4700503349304199, "mean_cosine": 0.76497483253479, "fve_rawvar": 0.35019635086254053, "learning_rate": 6.150880822748424e-07, "grad_norm": 2.068732738494873, "pred_norm_raw": 114.91481018066406, "gold_norm_raw": 120.57661437988281, "backbone_norm_raw": 114.68305969238281, "elapsed_seconds": 6000.14404630661, "seconds_per_optimizer_step": 29.126912846148592} | |
| {"kind": "train", "optimizer_step": 207, "micro_step": 207, "loss": 0.4351677894592285, "mean_cosine": 0.7824161052703857, "fve_rawvar": 0.3984184318908177, "learning_rate": 6.133726235465173e-07, "grad_norm": 1.7897733449935913, "pred_norm_raw": 113.48111724853516, "gold_norm_raw": 119.8291244506836, "backbone_norm_raw": 113.06682586669922, "elapsed_seconds": 6025.697820901871, "seconds_per_optimizer_step": 29.109651308704688} | |
| {"kind": "train", "optimizer_step": 208, "micro_step": 208, "loss": 0.4838474988937378, "mean_cosine": 0.7580762505531311, "fve_rawvar": 0.331122973343431, "learning_rate": 6.123431772683718e-07, "grad_norm": 1.6613216400146484, "pred_norm_raw": 114.15825653076172, "gold_norm_raw": 116.13520050048828, "backbone_norm_raw": 113.80439758300781, "elapsed_seconds": 6051.142619371414, "seconds_per_optimizer_step": 29.09203182390103} | |
| {"kind": "train", "optimizer_step": 209, "micro_step": 209, "loss": 0.386888325214386, "mean_cosine": 0.806555837392807, "fve_rawvar": 0.46516058632273427, "learning_rate": 6.12e-07, "grad_norm": 1.843657374382019, "pred_norm_raw": 114.82235717773438, "gold_norm_raw": 121.4876708984375, "backbone_norm_raw": 114.39256286621094, "elapsed_seconds": 6076.575217008591, "seconds_per_optimizer_step": 29.074522569419095} | |
| {"kind": "eval", "optimizer_step": 209, "real_loss": 0.47344916748503846, "shuffled_loss": 1.2868549211157694, "conditioning_gap": 0.8134057536307309, "mean_cosine": 0.7632754162574807, "fve_rawvar": 0.3408344631438017, "eval_seconds": 37.785274267196655, "elapsed_seconds": 6114.371411561966} | |
| {"kind": "checkpoint", "optimizer_step": 209, "path": "/home/sparrow/NLAmodel/experiments/Exp18_affect_ar_sft/runs/ar5000_3gpu/checkpoints/step_000209", "elapsed_seconds": 6114.380260467529} | |
Xet Storage Details
- Size:
- 93.5 kB
- Xet hash:
- 9fd5db395d47ce8ddc07da03ad21970b027c11bd8fe48edd7993099916c134a2
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.