| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.5798486658701713, |
| "eval_steps": 500, |
| "global_step": 364, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.00796495420151334, |
| "grad_norm": 0.31925168633461, |
| "learning_rate": 1.8181818181818182e-05, |
| "loss": 0.2621, |
| "mean_token_accuracy": 0.9140689849853516, |
| "num_tokens": 976840.0, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.01592990840302668, |
| "grad_norm": 0.08322103321552277, |
| "learning_rate": 4.0909090909090915e-05, |
| "loss": 0.1972, |
| "mean_token_accuracy": 0.9554547607898712, |
| "num_tokens": 1946985.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.023894862604540025, |
| "grad_norm": 0.12275099009275436, |
| "learning_rate": 6.363636363636364e-05, |
| "loss": 0.1185, |
| "mean_token_accuracy": 0.9702158480882644, |
| "num_tokens": 2900082.0, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.03185981680605336, |
| "grad_norm": 0.7107292413711548, |
| "learning_rate": 8.636363636363637e-05, |
| "loss": 0.1282, |
| "mean_token_accuracy": 0.9664572983980179, |
| "num_tokens": 3864699.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.039824771007566706, |
| "grad_norm": 0.23207606375217438, |
| "learning_rate": 9.999951389537729e-05, |
| "loss": 0.2217, |
| "mean_token_accuracy": 0.9509527534246445, |
| "num_tokens": 4828617.0, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.04778972520908005, |
| "grad_norm": 0.14495718479156494, |
| "learning_rate": 9.999404536310313e-05, |
| "loss": 0.2064, |
| "mean_token_accuracy": 0.9512994945049286, |
| "num_tokens": 5787040.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.05575467941059339, |
| "grad_norm": 0.12353721261024475, |
| "learning_rate": 9.998250155681299e-05, |
| "loss": 0.1701, |
| "mean_token_accuracy": 0.9584007978439331, |
| "num_tokens": 6752619.0, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.06371963361210672, |
| "grad_norm": 0.21419960260391235, |
| "learning_rate": 9.99648843469849e-05, |
| "loss": 0.1511, |
| "mean_token_accuracy": 0.9626434803009033, |
| "num_tokens": 7726001.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.07168458781362007, |
| "grad_norm": 3.3304569721221924, |
| "learning_rate": 9.994119658818884e-05, |
| "loss": 0.2765, |
| "mean_token_accuracy": 0.9528724819421768, |
| "num_tokens": 8685074.0, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.07964954201513341, |
| "grad_norm": 0.05958124250173569, |
| "learning_rate": 9.991144211862432e-05, |
| "loss": 0.154, |
| "mean_token_accuracy": 0.9638805568218232, |
| "num_tokens": 9655550.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08761449621664676, |
| "grad_norm": 0.22341704368591309, |
| "learning_rate": 9.987562575949834e-05, |
| "loss": 0.1403, |
| "mean_token_accuracy": 0.9647653281688691, |
| "num_tokens": 10617273.0, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.0955794504181601, |
| "grad_norm": 0.057139936834573746, |
| "learning_rate": 9.983375331424436e-05, |
| "loss": 0.0785, |
| "mean_token_accuracy": 0.9757441192865371, |
| "num_tokens": 11579272.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.10354440461967343, |
| "grad_norm": 0.04204887896776199, |
| "learning_rate": 9.978583156758181e-05, |
| "loss": 0.0999, |
| "mean_token_accuracy": 0.9719029635190963, |
| "num_tokens": 12545761.0, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.11150935882118677, |
| "grad_norm": 0.16803434491157532, |
| "learning_rate": 9.973186828441682e-05, |
| "loss": 0.0845, |
| "mean_token_accuracy": 0.9757595360279083, |
| "num_tokens": 13496986.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.11947431302270012, |
| "grad_norm": 0.0414872020483017, |
| "learning_rate": 9.967187220858398e-05, |
| "loss": 0.0985, |
| "mean_token_accuracy": 0.9712514847517013, |
| "num_tokens": 14470911.0, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.12743926722421345, |
| "grad_norm": 0.16572175920009613, |
| "learning_rate": 9.960585306142969e-05, |
| "loss": 0.1053, |
| "mean_token_accuracy": 0.9700259149074555, |
| "num_tokens": 15431914.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.1354042214257268, |
| "grad_norm": 0.14159245789051056, |
| "learning_rate": 9.953382154023676e-05, |
| "loss": 0.1029, |
| "mean_token_accuracy": 0.9705437779426574, |
| "num_tokens": 16399339.0, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.14336917562724014, |
| "grad_norm": 0.16658474504947662, |
| "learning_rate": 9.945578931649133e-05, |
| "loss": 0.1792, |
| "mean_token_accuracy": 0.9554636508226395, |
| "num_tokens": 17355529.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.1513341298287535, |
| "grad_norm": 0.1064157709479332, |
| "learning_rate": 9.937176903399153e-05, |
| "loss": 0.1245, |
| "mean_token_accuracy": 0.9657773017883301, |
| "num_tokens": 18326666.0, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.15929908403026682, |
| "grad_norm": 0.038611143827438354, |
| "learning_rate": 9.928177430679881e-05, |
| "loss": 0.0554, |
| "mean_token_accuracy": 0.9804502189159393, |
| "num_tokens": 19271602.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16726403823178015, |
| "grad_norm": 0.055337049067020416, |
| "learning_rate": 9.918581971703215e-05, |
| "loss": 0.1013, |
| "mean_token_accuracy": 0.9726076275110245, |
| "num_tokens": 20239228.0, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.1752289924332935, |
| "grad_norm": 0.07450346648693085, |
| "learning_rate": 9.908392081250502e-05, |
| "loss": 0.1557, |
| "mean_token_accuracy": 0.9616903692483902, |
| "num_tokens": 21193278.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.18319394663480684, |
| "grad_norm": 0.05388251692056656, |
| "learning_rate": 9.897609410420636e-05, |
| "loss": 0.0579, |
| "mean_token_accuracy": 0.9801422923803329, |
| "num_tokens": 22157071.0, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.1911589008363202, |
| "grad_norm": 0.14010213315486908, |
| "learning_rate": 9.886235706362512e-05, |
| "loss": 0.129, |
| "mean_token_accuracy": 0.9660608291625976, |
| "num_tokens": 23126800.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.19912385503783353, |
| "grad_norm": 0.06920307874679565, |
| "learning_rate": 9.87427281199193e-05, |
| "loss": 0.0642, |
| "mean_token_accuracy": 0.9792020440101623, |
| "num_tokens": 24078851.0, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.20708880923934686, |
| "grad_norm": 0.16701486706733704, |
| "learning_rate": 9.86172266569299e-05, |
| "loss": 0.1406, |
| "mean_token_accuracy": 0.9666702151298523, |
| "num_tokens": 25037639.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.21505376344086022, |
| "grad_norm": 0.07498443126678467, |
| "learning_rate": 9.848587301004004e-05, |
| "loss": 0.0777, |
| "mean_token_accuracy": 0.9759969502687454, |
| "num_tokens": 25998711.0, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.22301871764237355, |
| "grad_norm": 0.04914296790957451, |
| "learning_rate": 9.834868846287998e-05, |
| "loss": 0.0795, |
| "mean_token_accuracy": 0.9764232695102691, |
| "num_tokens": 26959256.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.2309836718438869, |
| "grad_norm": 0.19421175122261047, |
| "learning_rate": 9.820569524387842e-05, |
| "loss": 0.1555, |
| "mean_token_accuracy": 0.9619136154651642, |
| "num_tokens": 27939252.0, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.23894862604540024, |
| "grad_norm": 0.20248696208000183, |
| "learning_rate": 9.805691652266081e-05, |
| "loss": 0.1381, |
| "mean_token_accuracy": 0.967031866312027, |
| "num_tokens": 28904359.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.24691358024691357, |
| "grad_norm": 0.04761936515569687, |
| "learning_rate": 9.790237640629512e-05, |
| "loss": 0.0813, |
| "mean_token_accuracy": 0.9761160612106323, |
| "num_tokens": 29855029.0, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.2548785344484269, |
| "grad_norm": 0.15918390452861786, |
| "learning_rate": 9.774209993538562e-05, |
| "loss": 0.1392, |
| "mean_token_accuracy": 0.9662097334861756, |
| "num_tokens": 30811383.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.2628434886499403, |
| "grad_norm": 0.11582732200622559, |
| "learning_rate": 9.757611308001557e-05, |
| "loss": 0.1058, |
| "mean_token_accuracy": 0.9722520887851716, |
| "num_tokens": 31781550.0, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.2708084428514536, |
| "grad_norm": 0.0649627223610878, |
| "learning_rate": 9.740444273553913e-05, |
| "loss": 0.077, |
| "mean_token_accuracy": 0.9758875638246536, |
| "num_tokens": 32755034.0, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.27877339705296694, |
| "grad_norm": 0.04820963740348816, |
| "learning_rate": 9.722711671822347e-05, |
| "loss": 0.1283, |
| "mean_token_accuracy": 0.9675563126802444, |
| "num_tokens": 33735769.0, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.2867383512544803, |
| "grad_norm": 0.14796513319015503, |
| "learning_rate": 9.704416376074162e-05, |
| "loss": 0.1181, |
| "mean_token_accuracy": 0.9659722477197648, |
| "num_tokens": 34705651.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.2947033054559936, |
| "grad_norm": 0.07465963810682297, |
| "learning_rate": 9.685561350751678e-05, |
| "loss": 0.0658, |
| "mean_token_accuracy": 0.9791929066181183, |
| "num_tokens": 35673728.0, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.302668259657507, |
| "grad_norm": 0.2720688581466675, |
| "learning_rate": 9.666149650991898e-05, |
| "loss": 0.2147, |
| "mean_token_accuracy": 0.946672010421753, |
| "num_tokens": 36646764.0, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.3106332138590203, |
| "grad_norm": 0.09100554883480072, |
| "learning_rate": 9.646184422131477e-05, |
| "loss": 0.0605, |
| "mean_token_accuracy": 0.9802939683198929, |
| "num_tokens": 37607102.0, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.31859816806053365, |
| "grad_norm": 0.1170743927359581, |
| "learning_rate": 9.625668899197068e-05, |
| "loss": 0.0909, |
| "mean_token_accuracy": 0.9745945423841477, |
| "num_tokens": 38572907.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.326563122262047, |
| "grad_norm": 0.18812328577041626, |
| "learning_rate": 9.604606406381143e-05, |
| "loss": 0.1636, |
| "mean_token_accuracy": 0.9610334515571595, |
| "num_tokens": 39522511.0, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.3345280764635603, |
| "grad_norm": 0.07258269190788269, |
| "learning_rate": 9.583000356503371e-05, |
| "loss": 0.0981, |
| "mean_token_accuracy": 0.9744321316480636, |
| "num_tokens": 40490815.0, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.3424930306650737, |
| "grad_norm": 0.134218230843544, |
| "learning_rate": 9.560854250457614e-05, |
| "loss": 0.0912, |
| "mean_token_accuracy": 0.9722292870283127, |
| "num_tokens": 41444929.0, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.350457984866587, |
| "grad_norm": 0.0713258907198906, |
| "learning_rate": 9.538171676644679e-05, |
| "loss": 0.0754, |
| "mean_token_accuracy": 0.9765953630208969, |
| "num_tokens": 42421307.0, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.35842293906810035, |
| "grad_norm": 0.05106978863477707, |
| "learning_rate": 9.514956310390874e-05, |
| "loss": 0.0869, |
| "mean_token_accuracy": 0.9751714557409287, |
| "num_tokens": 43382180.0, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.3663878932696137, |
| "grad_norm": 0.03917595371603966, |
| "learning_rate": 9.491211913352484e-05, |
| "loss": 0.0899, |
| "mean_token_accuracy": 0.9736981391906738, |
| "num_tokens": 44332878.0, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.374352847471127, |
| "grad_norm": 0.10567553341388702, |
| "learning_rate": 9.466942332906257e-05, |
| "loss": 0.1384, |
| "mean_token_accuracy": 0.9642836421728134, |
| "num_tokens": 45316199.0, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.3823178016726404, |
| "grad_norm": 0.04932386800646782, |
| "learning_rate": 9.442151501526004e-05, |
| "loss": 0.0825, |
| "mean_token_accuracy": 0.976224085688591, |
| "num_tokens": 46290796.0, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.39028275587415373, |
| "grad_norm": 0.10389234870672226, |
| "learning_rate": 9.416843436145408e-05, |
| "loss": 0.0768, |
| "mean_token_accuracy": 0.9771006613969803, |
| "num_tokens": 47260011.0, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.39824771007566706, |
| "grad_norm": 0.07096272706985474, |
| "learning_rate": 9.391022237507148e-05, |
| "loss": 0.1473, |
| "mean_token_accuracy": 0.9629392325878143, |
| "num_tokens": 48227591.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.4062126642771804, |
| "grad_norm": 0.039991188794374466, |
| "learning_rate": 9.364692089498442e-05, |
| "loss": 0.1001, |
| "mean_token_accuracy": 0.9721494376659393, |
| "num_tokens": 49176834.0, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.4141776184786937, |
| "grad_norm": 0.059157680720090866, |
| "learning_rate": 9.337857258473119e-05, |
| "loss": 0.0836, |
| "mean_token_accuracy": 0.9765387207269669, |
| "num_tokens": 50128221.0, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.4221425726802071, |
| "grad_norm": 0.07804446667432785, |
| "learning_rate": 9.31052209256033e-05, |
| "loss": 0.0804, |
| "mean_token_accuracy": 0.9744845539331436, |
| "num_tokens": 51106738.0, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.43010752688172044, |
| "grad_norm": 0.06593296676874161, |
| "learning_rate": 9.282691020960008e-05, |
| "loss": 0.0656, |
| "mean_token_accuracy": 0.9785562753677368, |
| "num_tokens": 52078332.0, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.43807248108323377, |
| "grad_norm": 0.16336949169635773, |
| "learning_rate": 9.254368553225191e-05, |
| "loss": 0.0827, |
| "mean_token_accuracy": 0.9763776600360871, |
| "num_tokens": 53040460.0, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.4460374352847471, |
| "grad_norm": 0.06217002868652344, |
| "learning_rate": 9.225559278531323e-05, |
| "loss": 0.0747, |
| "mean_token_accuracy": 0.9773959875106811, |
| "num_tokens": 53998407.0, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.4540023894862604, |
| "grad_norm": 0.18053705990314484, |
| "learning_rate": 9.196267864932657e-05, |
| "loss": 0.0973, |
| "mean_token_accuracy": 0.9719738304615021, |
| "num_tokens": 54957483.0, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.4619673436877738, |
| "grad_norm": 0.04883285239338875, |
| "learning_rate": 9.16649905860588e-05, |
| "loss": 0.0905, |
| "mean_token_accuracy": 0.9736559271812439, |
| "num_tokens": 55910394.0, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.46993229788928714, |
| "grad_norm": 0.054185543209314346, |
| "learning_rate": 9.136257683081068e-05, |
| "loss": 0.0712, |
| "mean_token_accuracy": 0.9781385570764541, |
| "num_tokens": 56889841.0, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.4778972520908005, |
| "grad_norm": 0.1914336234331131, |
| "learning_rate": 9.105548638460124e-05, |
| "loss": 0.0986, |
| "mean_token_accuracy": 0.972733274102211, |
| "num_tokens": 57864522.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.4858622062923138, |
| "grad_norm": 0.24836596846580505, |
| "learning_rate": 9.074376900622789e-05, |
| "loss": 0.0948, |
| "mean_token_accuracy": 0.9740723758935929, |
| "num_tokens": 58839816.0, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.49382716049382713, |
| "grad_norm": 0.20251403748989105, |
| "learning_rate": 9.042747520420394e-05, |
| "loss": 0.1383, |
| "mean_token_accuracy": 0.9654656767845153, |
| "num_tokens": 59806160.0, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.5017921146953405, |
| "grad_norm": 0.15884502232074738, |
| "learning_rate": 9.010665622857454e-05, |
| "loss": 0.1026, |
| "mean_token_accuracy": 0.972476851940155, |
| "num_tokens": 60770582.0, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.5097570688968538, |
| "grad_norm": 0.1830815225839615, |
| "learning_rate": 8.978136406261241e-05, |
| "loss": 0.116, |
| "mean_token_accuracy": 0.9693456411361694, |
| "num_tokens": 61755317.0, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.5177220230983672, |
| "grad_norm": 0.14333826303482056, |
| "learning_rate": 8.945165141439494e-05, |
| "loss": 0.0959, |
| "mean_token_accuracy": 0.9719290375709534, |
| "num_tokens": 62731764.0, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.5256869772998806, |
| "grad_norm": 0.05398755893111229, |
| "learning_rate": 8.911757170826361e-05, |
| "loss": 0.0696, |
| "mean_token_accuracy": 0.97956962287426, |
| "num_tokens": 63679572.0, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.5336519315013939, |
| "grad_norm": 0.15619242191314697, |
| "learning_rate": 8.877917907616764e-05, |
| "loss": 0.1589, |
| "mean_token_accuracy": 0.960988137125969, |
| "num_tokens": 64655346.0, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.5416168857029072, |
| "grad_norm": 0.049750592559576035, |
| "learning_rate": 8.843652834889265e-05, |
| "loss": 0.0919, |
| "mean_token_accuracy": 0.972157496213913, |
| "num_tokens": 65632780.0, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.5495818399044206, |
| "grad_norm": 0.06751636415719986, |
| "learning_rate": 8.808967504717642e-05, |
| "loss": 0.0777, |
| "mean_token_accuracy": 0.9761724084615707, |
| "num_tokens": 66597305.0, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.5575467941059339, |
| "grad_norm": 0.05424492433667183, |
| "learning_rate": 8.773867537271258e-05, |
| "loss": 0.1091, |
| "mean_token_accuracy": 0.9714101910591125, |
| "num_tokens": 67559552.0, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.5655117483074472, |
| "grad_norm": 0.173378124833107, |
| "learning_rate": 8.738358619904417e-05, |
| "loss": 0.0657, |
| "mean_token_accuracy": 0.980780491232872, |
| "num_tokens": 68516882.0, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.5734767025089605, |
| "grad_norm": 0.24044422805309296, |
| "learning_rate": 8.702446506234817e-05, |
| "loss": 0.137, |
| "mean_token_accuracy": 0.9668047279119492, |
| "num_tokens": 69470177.0, |
| "step": 360 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 1256, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 5.599814568445317e+17, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
