| [ |
| { |
| "loss": 1.6756, |
| "grad_norm": 0.9338335990905762, |
| "learning_rate": 1.8e-05, |
| "entropy": 1.2653020560741424, |
| "num_tokens": 211745.0, |
| "mean_token_accuracy": 0.626320019364357, |
| "epoch": 0.016, |
| "step": 10 |
| }, |
| { |
| "loss": 1.6675, |
| "grad_norm": 0.4600104093551636, |
| "learning_rate": 3.8e-05, |
| "entropy": 1.5156079709529877, |
| "num_tokens": 413783.0, |
| "mean_token_accuracy": 0.6156355381011963, |
| "epoch": 0.032, |
| "step": 20 |
| }, |
| { |
| "loss": 1.4313, |
| "grad_norm": 0.5679482221603394, |
| "learning_rate": 5.8e-05, |
| "entropy": 1.511820811033249, |
| "num_tokens": 618078.0, |
| "mean_token_accuracy": 0.6563311725854873, |
| "epoch": 0.048, |
| "step": 30 |
| }, |
| { |
| "loss": 1.1182, |
| "grad_norm": 0.5529033541679382, |
| "learning_rate": 7.800000000000001e-05, |
| "entropy": 1.1426804274320603, |
| "num_tokens": 824579.0, |
| "mean_token_accuracy": 0.7136697113513947, |
| "epoch": 0.064, |
| "step": 40 |
| }, |
| { |
| "loss": 0.9574, |
| "grad_norm": 0.4290783405303955, |
| "learning_rate": 9.8e-05, |
| "entropy": 0.9684652209281921, |
| "num_tokens": 1022359.0, |
| "mean_token_accuracy": 0.7489378124475479, |
| "epoch": 0.08, |
| "step": 50 |
| }, |
| { |
| "loss": 0.8724, |
| "grad_norm": 0.34012866020202637, |
| "learning_rate": 9.993956318446873e-05, |
| "entropy": 0.8785539418458939, |
| "num_tokens": 1237697.0, |
| "mean_token_accuracy": 0.7662121266126632, |
| "epoch": 0.096, |
| "step": 60 |
| }, |
| { |
| "loss": 0.8526, |
| "grad_norm": 0.4122697710990906, |
| "learning_rate": 9.973083336646172e-05, |
| "entropy": 0.8518908679485321, |
| "num_tokens": 1438438.0, |
| "mean_token_accuracy": 0.7753641337156296, |
| "epoch": 0.112, |
| "step": 70 |
| }, |
| { |
| "loss": 0.7437, |
| "grad_norm": 0.3228837251663208, |
| "learning_rate": 9.937368719576374e-05, |
| "entropy": 0.7489307045936584, |
| "num_tokens": 1631589.0, |
| "mean_token_accuracy": 0.7941762745380402, |
| "epoch": 0.128, |
| "step": 80 |
| }, |
| { |
| "loss": 0.7492, |
| "grad_norm": 0.33696669340133667, |
| "learning_rate": 9.886919053691883e-05, |
| "entropy": 0.7552473366260528, |
| "num_tokens": 1836258.0, |
| "mean_token_accuracy": 0.7946388304233551, |
| "epoch": 0.144, |
| "step": 90 |
| }, |
| { |
| "loss": 0.7372, |
| "grad_norm": 0.2882266640663147, |
| "learning_rate": 9.82188490062415e-05, |
| "entropy": 0.7464645385742188, |
| "num_tokens": 2046902.0, |
| "mean_token_accuracy": 0.7958032578229904, |
| "epoch": 0.16, |
| "step": 100 |
| }, |
| { |
| "loss": 0.7528, |
| "grad_norm": 0.34871771931648254, |
| "learning_rate": 9.742460347846587e-05, |
| "entropy": 0.7529336363077164, |
| "num_tokens": 2254427.0, |
| "mean_token_accuracy": 0.7927632719278336, |
| "epoch": 0.176, |
| "step": 110 |
| }, |
| { |
| "loss": 0.6854, |
| "grad_norm": 0.344471275806427, |
| "learning_rate": 9.648882429441257e-05, |
| "entropy": 0.6884129703044891, |
| "num_tokens": 2452097.0, |
| "mean_token_accuracy": 0.8083581000566482, |
| "epoch": 0.192, |
| "step": 120 |
| }, |
| { |
| "loss": 0.7035, |
| "grad_norm": 0.2790994942188263, |
| "learning_rate": 9.541430418696018e-05, |
| "entropy": 0.712130931019783, |
| "num_tokens": 2647959.0, |
| "mean_token_accuracy": 0.8047784239053726, |
| "epoch": 0.208, |
| "step": 130 |
| }, |
| { |
| "loss": 0.7175, |
| "grad_norm": 0.3251378536224365, |
| "learning_rate": 9.420424994643266e-05, |
| "entropy": 0.7225965172052383, |
| "num_tokens": 2851644.0, |
| "mean_token_accuracy": 0.8003872334957123, |
| "epoch": 0.224, |
| "step": 140 |
| }, |
| { |
| "loss": 0.6633, |
| "grad_norm": 0.3178017735481262, |
| "learning_rate": 9.28622728502766e-05, |
| "entropy": 0.6729940563440323, |
| "num_tokens": 3053179.0, |
| "mean_token_accuracy": 0.8133321523666381, |
| "epoch": 0.24, |
| "step": 150 |
| }, |
| { |
| "loss": 0.7078, |
| "grad_norm": 0.32298043370246887, |
| "learning_rate": 9.139237788559e-05, |
| "entropy": 0.716013514995575, |
| "num_tokens": 3244569.0, |
| "mean_token_accuracy": 0.8037430495023727, |
| "epoch": 0.256, |
| "step": 160 |
| }, |
| { |
| "loss": 0.7011, |
| "grad_norm": 0.3123107850551605, |
| "learning_rate": 8.979895179666673e-05, |
| "entropy": 0.6998003959655762, |
| "num_tokens": 3448693.0, |
| "mean_token_accuracy": 0.8040744125843048, |
| "epoch": 0.272, |
| "step": 170 |
| }, |
| { |
| "loss": 0.6887, |
| "grad_norm": 0.3589050769805908, |
| "learning_rate": 8.808674999322728e-05, |
| "entropy": 0.6893662005662918, |
| "num_tokens": 3650299.0, |
| "mean_token_accuracy": 0.8075590372085572, |
| "epoch": 0.288, |
| "step": 180 |
| }, |
| { |
| "loss": 0.7397, |
| "grad_norm": 0.3437199890613556, |
| "learning_rate": 8.626088235840726e-05, |
| "entropy": 0.7490889757871628, |
| "num_tokens": 3860706.0, |
| "mean_token_accuracy": 0.7960624009370804, |
| "epoch": 0.304, |
| "step": 190 |
| }, |
| { |
| "loss": 0.7104, |
| "grad_norm": 0.31088024377822876, |
| "learning_rate": 8.43267979988576e-05, |
| "entropy": 0.7115123689174652, |
| "num_tokens": 4057498.0, |
| "mean_token_accuracy": 0.8006591260433197, |
| "epoch": 0.32, |
| "step": 200 |
| }, |
| { |
| "loss": 0.6628, |
| "grad_norm": 0.3589341640472412, |
| "learning_rate": 8.229026898246885e-05, |
| "entropy": 0.6689569860696792, |
| "num_tokens": 4251824.0, |
| "mean_token_accuracy": 0.812098690867424, |
| "epoch": 0.336, |
| "step": 210 |
| }, |
| { |
| "loss": 0.676, |
| "grad_norm": 0.31112033128738403, |
| "learning_rate": 8.015737311225172e-05, |
| "entropy": 0.6726008355617523, |
| "num_tokens": 4465389.0, |
| "mean_token_accuracy": 0.8103758335113526, |
| "epoch": 0.352, |
| "step": 220 |
| }, |
| { |
| "loss": 0.6668, |
| "grad_norm": 0.33138927817344666, |
| "learning_rate": 7.793447578778427e-05, |
| "entropy": 0.6735879182815552, |
| "num_tokens": 4685518.0, |
| "mean_token_accuracy": 0.8112154036760331, |
| "epoch": 0.368, |
| "step": 230 |
| }, |
| { |
| "loss": 0.6414, |
| "grad_norm": 0.31951209902763367, |
| "learning_rate": 7.56282110083577e-05, |
| "entropy": 0.6442272961139679, |
| "num_tokens": 4886383.0, |
| "mean_token_accuracy": 0.8160840034484863, |
| "epoch": 0.384, |
| "step": 240 |
| }, |
| { |
| "loss": 0.6817, |
| "grad_norm": 0.3405151665210724, |
| "learning_rate": 7.324546157451508e-05, |
| "entropy": 0.6818776607513428, |
| "num_tokens": 5086529.0, |
| "mean_token_accuracy": 0.8091985791921615, |
| "epoch": 0.4, |
| "step": 250 |
| }, |
| { |
| "loss": 0.6843, |
| "grad_norm": 0.30915334820747375, |
| "learning_rate": 7.079333854706938e-05, |
| "entropy": 0.6939111232757569, |
| "num_tokens": 5298416.0, |
| "mean_token_accuracy": 0.8071506917476654, |
| "epoch": 0.416, |
| "step": 260 |
| }, |
| { |
| "loss": 0.6805, |
| "grad_norm": 0.3196154832839966, |
| "learning_rate": 6.827916002490304e-05, |
| "entropy": 0.6853002905845642, |
| "num_tokens": 5502693.0, |
| "mean_token_accuracy": 0.8068861842155457, |
| "epoch": 0.432, |
| "step": 270 |
| }, |
| { |
| "loss": 0.6497, |
| "grad_norm": 0.3364115059375763, |
| "learning_rate": 6.571042930488474e-05, |
| "entropy": 0.6549761325120926, |
| "num_tokens": 5697863.0, |
| "mean_token_accuracy": 0.81393081843853, |
| "epoch": 0.448, |
| "step": 280 |
| }, |
| { |
| "loss": 0.6697, |
| "grad_norm": 0.33196505904197693, |
| "learning_rate": 6.309481248908235e-05, |
| "entropy": 0.6747944802045822, |
| "num_tokens": 5910487.0, |
| "mean_token_accuracy": 0.8128439605236053, |
| "epoch": 0.464, |
| "step": 290 |
| }, |
| { |
| "loss": 0.6535, |
| "grad_norm": 0.30423638224601746, |
| "learning_rate": 6.044011560610153e-05, |
| "entropy": 0.6566255927085877, |
| "num_tokens": 6127477.0, |
| "mean_token_accuracy": 0.8121924549341202, |
| "epoch": 0.48, |
| "step": 300 |
| }, |
| { |
| "loss": 0.6518, |
| "grad_norm": 0.3487173914909363, |
| "learning_rate": 5.775426131482811e-05, |
| "entropy": 0.6593502879142761, |
| "num_tokens": 6328197.0, |
| "mean_token_accuracy": 0.814925542473793, |
| "epoch": 0.496, |
| "step": 310 |
| }, |
| { |
| "loss": 0.6931, |
| "grad_norm": 0.3142063021659851, |
| "learning_rate": 5.5045265260100234e-05, |
| "entropy": 0.700766509771347, |
| "num_tokens": 6536365.0, |
| "mean_token_accuracy": 0.8070295184850693, |
| "epoch": 0.512, |
| "step": 320 |
| }, |
| { |
| "loss": 0.6785, |
| "grad_norm": 0.3409480154514313, |
| "learning_rate": 5.232121215087369e-05, |
| "entropy": 0.6776748090982437, |
| "num_tokens": 6742650.0, |
| "mean_token_accuracy": 0.809091717004776, |
| "epoch": 0.528, |
| "step": 330 |
| }, |
| { |
| "loss": 0.6438, |
| "grad_norm": 0.3378758132457733, |
| "learning_rate": 4.959023163227284e-05, |
| "entropy": 0.6490969121456146, |
| "num_tokens": 6946868.0, |
| "mean_token_accuracy": 0.8178076684474945, |
| "epoch": 0.544, |
| "step": 340 |
| }, |
| { |
| "loss": 0.6481, |
| "grad_norm": 0.3258441388607025, |
| "learning_rate": 4.6860474023534335e-05, |
| "entropy": 0.6563442766666412, |
| "num_tokens": 7138705.0, |
| "mean_token_accuracy": 0.817442661523819, |
| "epoch": 0.56, |
| "step": 350 |
| }, |
| { |
| "loss": 0.6622, |
| "grad_norm": 0.32307952642440796, |
| "learning_rate": 4.4140085994251136e-05, |
| "entropy": 0.6632782518863678, |
| "num_tokens": 7348846.0, |
| "mean_token_accuracy": 0.812923377752304, |
| "epoch": 0.576, |
| "step": 360 |
| }, |
| { |
| "loss": 0.6436, |
| "grad_norm": 0.35711175203323364, |
| "learning_rate": 4.143718625150854e-05, |
| "entropy": 0.6440205127000809, |
| "num_tokens": 7556952.0, |
| "mean_token_accuracy": 0.8172848135232925, |
| "epoch": 0.592, |
| "step": 370 |
| }, |
| { |
| "loss": 0.6299, |
| "grad_norm": 0.3042103052139282, |
| "learning_rate": 3.875984131047135e-05, |
| "entropy": 0.6329021096229553, |
| "num_tokens": 7758339.0, |
| "mean_token_accuracy": 0.8194414287805557, |
| "epoch": 0.608, |
| "step": 380 |
| }, |
| { |
| "loss": 0.6559, |
| "grad_norm": 0.35361942648887634, |
| "learning_rate": 3.6116041420732103e-05, |
| "entropy": 0.6580730825662613, |
| "num_tokens": 7961705.0, |
| "mean_token_accuracy": 0.8145245909690857, |
| "epoch": 0.624, |
| "step": 390 |
| }, |
| { |
| "loss": 0.6394, |
| "grad_norm": 0.35071322321891785, |
| "learning_rate": 3.3513676720265934e-05, |
| "entropy": 0.6357898592948914, |
| "num_tokens": 8160317.0, |
| "mean_token_accuracy": 0.8177996665239334, |
| "epoch": 0.64, |
| "step": 400 |
| }, |
| { |
| "loss": 0.6545, |
| "grad_norm": 0.31700801849365234, |
| "learning_rate": 3.0960513688157754e-05, |
| "entropy": 0.649863663315773, |
| "num_tokens": 8372896.0, |
| "mean_token_accuracy": 0.8160782635211945, |
| "epoch": 0.656, |
| "step": 410 |
| }, |
| { |
| "loss": 0.656, |
| "grad_norm": 0.3056493401527405, |
| "learning_rate": 2.846417196637613e-05, |
| "entropy": 0.6630118876695633, |
| "num_tokens": 8582698.0, |
| "mean_token_accuracy": 0.8151620656251908, |
| "epoch": 0.672, |
| "step": 420 |
| }, |
| { |
| "loss": 0.6869, |
| "grad_norm": 0.3303000330924988, |
| "learning_rate": 2.603210161976687e-05, |
| "entropy": 0.6842911779880524, |
| "num_tokens": 8794332.0, |
| "mean_token_accuracy": 0.805791437625885, |
| "epoch": 0.688, |
| "step": 430 |
| }, |
| { |
| "loss": 0.6602, |
| "grad_norm": 0.3385516405105591, |
| "learning_rate": 2.3671560902131447e-05, |
| "entropy": 0.6720612704753876, |
| "num_tokens": 9008268.0, |
| "mean_token_accuracy": 0.811782768368721, |
| "epoch": 0.704, |
| "step": 440 |
| }, |
| { |
| "loss": 0.6425, |
| "grad_norm": 0.3051205277442932, |
| "learning_rate": 2.1389594594744864e-05, |
| "entropy": 0.6423384606838226, |
| "num_tokens": 9216146.0, |
| "mean_token_accuracy": 0.81755151450634, |
| "epoch": 0.72, |
| "step": 450 |
| }, |
| { |
| "loss": 0.6262, |
| "grad_norm": 0.3302038013935089, |
| "learning_rate": 1.9193012981959475e-05, |
| "entropy": 0.6281730026006699, |
| "num_tokens": 9418629.0, |
| "mean_token_accuracy": 0.820314571261406, |
| "epoch": 0.736, |
| "step": 460 |
| }, |
| { |
| "loss": 0.6204, |
| "grad_norm": 0.2998092472553253, |
| "learning_rate": 1.7088371526639853e-05, |
| "entropy": 0.6200591921806335, |
| "num_tokens": 9624430.0, |
| "mean_token_accuracy": 0.8230848222970962, |
| "epoch": 0.752, |
| "step": 470 |
| }, |
| { |
| "loss": 0.6133, |
| "grad_norm": 0.3248184025287628, |
| "learning_rate": 1.508195130608504e-05, |
| "entropy": 0.6175663053989411, |
| "num_tokens": 9824038.0, |
| "mean_token_accuracy": 0.8229408711194992, |
| "epoch": 0.768, |
| "step": 480 |
| }, |
| { |
| "loss": 0.6471, |
| "grad_norm": 0.3508250117301941, |
| "learning_rate": 1.3179740266825253e-05, |
| "entropy": 0.6427986294031143, |
| "num_tokens": 10029090.0, |
| "mean_token_accuracy": 0.8153195798397064, |
| "epoch": 0.784, |
| "step": 490 |
| }, |
| { |
| "loss": 0.6494, |
| "grad_norm": 0.3246026933193207, |
| "learning_rate": 1.1387415354235887e-05, |
| "entropy": 0.6497834831476211, |
| "num_tokens": 10233813.0, |
| "mean_token_accuracy": 0.8152256548404694, |
| "epoch": 0.8, |
| "step": 500 |
| }, |
| { |
| "loss": 0.621, |
| "grad_norm": 0.3171588182449341, |
| "learning_rate": 9.710325570301348e-06, |
| "entropy": 0.6244612663984299, |
| "num_tokens": 10435960.0, |
| "mean_token_accuracy": 0.8210752457380295, |
| "epoch": 0.816, |
| "step": 510 |
| }, |
| { |
| "loss": 0.5986, |
| "grad_norm": 0.32061895728111267, |
| "learning_rate": 8.153476010090788e-06, |
| "entropy": 0.5975286513566971, |
| "num_tokens": 10636030.0, |
| "mean_token_accuracy": 0.827234199643135, |
| "epoch": 0.832, |
| "step": 520 |
| }, |
| { |
| "loss": 0.6372, |
| "grad_norm": 0.3552604615688324, |
| "learning_rate": 6.721512924587242e-06, |
| "entropy": 0.6376388996839524, |
| "num_tokens": 10838906.0, |
| "mean_token_accuracy": 0.818820309638977, |
| "epoch": 0.848, |
| "step": 530 |
| }, |
| { |
| "loss": 0.6304, |
| "grad_norm": 0.33599594235420227, |
| "learning_rate": 5.418709854448489e-06, |
| "entropy": 0.6320249021053315, |
| "num_tokens": 11026457.0, |
| "mean_token_accuracy": 0.8198111385107041, |
| "epoch": 0.864, |
| "step": 540 |
| }, |
| { |
| "loss": 0.6235, |
| "grad_norm": 0.41409239172935486, |
| "learning_rate": 4.248954876082195e-06, |
| "entropy": 0.6273589804768562, |
| "num_tokens": 11211997.0, |
| "mean_token_accuracy": 0.8219926416873932, |
| "epoch": 0.88, |
| "step": 550 |
| }, |
| { |
| "loss": 0.6511, |
| "grad_norm": 0.337541401386261, |
| "learning_rate": 3.2157389980976783e-06, |
| "entropy": 0.6515613377094269, |
| "num_tokens": 11421486.0, |
| "mean_token_accuracy": 0.815014871954918, |
| "epoch": 0.896, |
| "step": 560 |
| }, |
| { |
| "loss": 0.6209, |
| "grad_norm": 0.3296094536781311, |
| "learning_rate": 2.322145742764181e-06, |
| "entropy": 0.6218553513288498, |
| "num_tokens": 11621269.0, |
| "mean_token_accuracy": 0.8230124861001968, |
| "epoch": 0.912, |
| "step": 570 |
| }, |
| { |
| "loss": 0.634, |
| "grad_norm": 0.34123408794403076, |
| "learning_rate": 1.5708419435684462e-06, |
| "entropy": 0.6355304539203643, |
| "num_tokens": 11830087.0, |
| "mean_token_accuracy": 0.818019899725914, |
| "epoch": 0.928, |
| "step": 580 |
| }, |
| { |
| "loss": 0.6276, |
| "grad_norm": 0.2921477258205414, |
| "learning_rate": 9.640697863354698e-07, |
| "entropy": 0.6308085203170777, |
| "num_tokens": 12030300.0, |
| "mean_token_accuracy": 0.8203556030988693, |
| "epoch": 0.944, |
| "step": 590 |
| }, |
| { |
| "loss": 0.623, |
| "grad_norm": 0.3348216712474823, |
| "learning_rate": 5.036401176647332e-07, |
| "entropy": 0.6214668571949005, |
| "num_tokens": 12226666.0, |
| "mean_token_accuracy": 0.8228818953037262, |
| "epoch": 0.96, |
| "step": 600 |
| }, |
| { |
| "loss": 0.6305, |
| "grad_norm": 0.3399432301521301, |
| "learning_rate": 1.9092704065222788e-07, |
| "entropy": 0.6307568103075027, |
| "num_tokens": 12433050.0, |
| "mean_token_accuracy": 0.8195924639701844, |
| "epoch": 0.976, |
| "step": 610 |
| }, |
| { |
| "loss": 0.6685, |
| "grad_norm": 0.3575892746448517, |
| "learning_rate": 2.6863814026734102e-08, |
| "entropy": 0.6648932933807373, |
| "num_tokens": 12635860.0, |
| "mean_token_accuracy": 0.8116440147161483, |
| "epoch": 0.992, |
| "step": 620 |
| }, |
| { |
| "train_runtime": 6933.2338, |
| "train_samples_per_second": 2.883, |
| "train_steps_per_second": 0.09, |
| "total_flos": 7.776202046473175e+18, |
| "train_loss": 0.7265998420715332, |
| "entropy": 0.6022609889507293, |
| "num_tokens": 12730954.0, |
| "mean_token_accuracy": 0.8287682414054871, |
| "epoch": 1.0, |
| "step": 625 |
| } |
| ] |
