| [ |
| { |
| "loss": 1.5327, |
| "grad_norm": 0.5706859827041626, |
| "learning_rate": 4.5e-05, |
| "entropy": 1.2147093176841737, |
| "num_tokens": 156309.0, |
| "mean_token_accuracy": 0.659215685725212, |
| "epoch": 0.03355704697986577, |
| "step": 10 |
| }, |
| { |
| "loss": 1.1979, |
| "grad_norm": 1.02685546875, |
| "learning_rate": 9.5e-05, |
| "entropy": 1.2502837777137756, |
| "num_tokens": 307512.0, |
| "mean_token_accuracy": 0.7190202981233597, |
| "epoch": 0.06711409395973154, |
| "step": 20 |
| }, |
| { |
| "loss": 0.6687, |
| "grad_norm": 0.5571576356887817, |
| "learning_rate": 9.999075138471951e-05, |
| "entropy": 0.6448976010084152, |
| "num_tokens": 461226.0, |
| "mean_token_accuracy": 0.8424648404121399, |
| "epoch": 0.10067114093959731, |
| "step": 30 |
| }, |
| { |
| "loss": 0.4812, |
| "grad_norm": 0.38987967371940613, |
| "learning_rate": 9.995878525539525e-05, |
| "entropy": 0.47405652403831483, |
| "num_tokens": 618054.0, |
| "mean_token_accuracy": 0.8851036787033081, |
| "epoch": 0.1342281879194631, |
| "step": 40 |
| }, |
| { |
| "loss": 0.3701, |
| "grad_norm": 0.2643287777900696, |
| "learning_rate": 9.990400202763563e-05, |
| "entropy": 0.3693989284336567, |
| "num_tokens": 775209.0, |
| "mean_token_accuracy": 0.9097828209400177, |
| "epoch": 0.16778523489932887, |
| "step": 50 |
| }, |
| { |
| "loss": 0.3193, |
| "grad_norm": 0.27373382449150085, |
| "learning_rate": 9.982642672195092e-05, |
| "entropy": 0.32156657576560976, |
| "num_tokens": 931365.0, |
| "mean_token_accuracy": 0.920613020658493, |
| "epoch": 0.20134228187919462, |
| "step": 60 |
| }, |
| { |
| "loss": 0.2866, |
| "grad_norm": 0.3094562292098999, |
| "learning_rate": 9.972609476841367e-05, |
| "entropy": 0.292499927431345, |
| "num_tokens": 1086695.0, |
| "mean_token_accuracy": 0.9282213032245636, |
| "epoch": 0.2348993288590604, |
| "step": 70 |
| }, |
| { |
| "loss": 0.2489, |
| "grad_norm": 0.2749999761581421, |
| "learning_rate": 9.960305199047712e-05, |
| "entropy": 0.2526983417570591, |
| "num_tokens": 1241470.0, |
| "mean_token_accuracy": 0.9362666577100753, |
| "epoch": 0.2684563758389262, |
| "step": 80 |
| }, |
| { |
| "loss": 0.2459, |
| "grad_norm": 0.2721326947212219, |
| "learning_rate": 9.945735458404681e-05, |
| "entropy": 0.24655402898788453, |
| "num_tokens": 1396709.0, |
| "mean_token_accuracy": 0.9367562144994735, |
| "epoch": 0.30201342281879195, |
| "step": 90 |
| }, |
| { |
| "loss": 0.2435, |
| "grad_norm": 0.21973158419132233, |
| "learning_rate": 9.928906909181481e-05, |
| "entropy": 0.24917888268828392, |
| "num_tokens": 1555025.0, |
| "mean_token_accuracy": 0.9364570289850235, |
| "epoch": 0.33557046979865773, |
| "step": 100 |
| }, |
| { |
| "loss": 0.2347, |
| "grad_norm": 0.26128071546554565, |
| "learning_rate": 9.909827237286849e-05, |
| "entropy": 0.23597020357847215, |
| "num_tokens": 1706763.0, |
| "mean_token_accuracy": 0.9403788715600967, |
| "epoch": 0.3691275167785235, |
| "step": 110 |
| }, |
| { |
| "loss": 0.2478, |
| "grad_norm": 0.20566503703594208, |
| "learning_rate": 9.888505156758759e-05, |
| "entropy": 0.2498910292983055, |
| "num_tokens": 1867753.0, |
| "mean_token_accuracy": 0.9367677628993988, |
| "epoch": 0.40268456375838924, |
| "step": 120 |
| }, |
| { |
| "loss": 0.2417, |
| "grad_norm": 0.190927654504776, |
| "learning_rate": 9.864950405784551e-05, |
| "entropy": 0.24509735256433487, |
| "num_tokens": 2027436.0, |
| "mean_token_accuracy": 0.9376339435577392, |
| "epoch": 0.436241610738255, |
| "step": 130 |
| }, |
| { |
| "loss": 0.2196, |
| "grad_norm": 0.1996077448129654, |
| "learning_rate": 9.839173742253334e-05, |
| "entropy": 0.22244628816843032, |
| "num_tokens": 2182978.0, |
| "mean_token_accuracy": 0.9424097180366516, |
| "epoch": 0.4697986577181208, |
| "step": 140 |
| }, |
| { |
| "loss": 0.2534, |
| "grad_norm": 0.21034187078475952, |
| "learning_rate": 9.811186938842645e-05, |
| "entropy": 0.2574092894792557, |
| "num_tokens": 2343108.0, |
| "mean_token_accuracy": 0.9341461777687072, |
| "epoch": 0.5033557046979866, |
| "step": 150 |
| }, |
| { |
| "loss": 0.2369, |
| "grad_norm": 0.20116521418094635, |
| "learning_rate": 9.781002777641664e-05, |
| "entropy": 0.23710015863180162, |
| "num_tokens": 2503180.0, |
| "mean_token_accuracy": 0.9385264277458191, |
| "epoch": 0.5369127516778524, |
| "step": 160 |
| }, |
| { |
| "loss": 0.2194, |
| "grad_norm": 0.20917288959026337, |
| "learning_rate": 9.748635044313386e-05, |
| "entropy": 0.22300637513399124, |
| "num_tokens": 2658840.0, |
| "mean_token_accuracy": 0.9410306721925735, |
| "epoch": 0.5704697986577181, |
| "step": 170 |
| }, |
| { |
| "loss": 0.2295, |
| "grad_norm": 0.19810253381729126, |
| "learning_rate": 9.714098521798465e-05, |
| "entropy": 0.23066828176379203, |
| "num_tokens": 2818656.0, |
| "mean_token_accuracy": 0.9388932436704636, |
| "epoch": 0.6040268456375839, |
| "step": 180 |
| }, |
| { |
| "loss": 0.231, |
| "grad_norm": 0.17845278978347778, |
| "learning_rate": 9.677408983563565e-05, |
| "entropy": 0.23543039262294768, |
| "num_tokens": 2975473.0, |
| "mean_token_accuracy": 0.9391510993242264, |
| "epoch": 0.6375838926174496, |
| "step": 190 |
| }, |
| { |
| "loss": 0.2152, |
| "grad_norm": 0.23437312245368958, |
| "learning_rate": 9.638583186397331e-05, |
| "entropy": 0.21993101984262467, |
| "num_tokens": 3130452.0, |
| "mean_token_accuracy": 0.9423492521047592, |
| "epoch": 0.6711409395973155, |
| "step": 200 |
| }, |
| { |
| "loss": 0.204, |
| "grad_norm": 0.2377660572528839, |
| "learning_rate": 9.597638862757255e-05, |
| "entropy": 0.2064872995018959, |
| "num_tokens": 3287490.0, |
| "mean_token_accuracy": 0.9451317518949509, |
| "epoch": 0.7046979865771812, |
| "step": 210 |
| }, |
| { |
| "loss": 0.2158, |
| "grad_norm": 0.17862877249717712, |
| "learning_rate": 9.554594712670926e-05, |
| "entropy": 0.2166791968047619, |
| "num_tokens": 3446266.0, |
| "mean_token_accuracy": 0.9421544283628464, |
| "epoch": 0.738255033557047, |
| "step": 220 |
| }, |
| { |
| "loss": 0.2253, |
| "grad_norm": 0.1511947214603424, |
| "learning_rate": 9.509470395195399e-05, |
| "entropy": 0.22676953300833702, |
| "num_tokens": 3602351.0, |
| "mean_token_accuracy": 0.939826962351799, |
| "epoch": 0.7718120805369127, |
| "step": 230 |
| }, |
| { |
| "loss": 0.2201, |
| "grad_norm": 0.26649826765060425, |
| "learning_rate": 9.46228651943853e-05, |
| "entropy": 0.22365730181336402, |
| "num_tokens": 3757932.0, |
| "mean_token_accuracy": 0.9410541921854019, |
| "epoch": 0.8053691275167785, |
| "step": 240 |
| }, |
| { |
| "loss": 0.2026, |
| "grad_norm": 0.19634084403514862, |
| "learning_rate": 9.413064635146418e-05, |
| "entropy": 0.20426617376506329, |
| "num_tokens": 3912346.0, |
| "mean_token_accuracy": 0.9445015460252761, |
| "epoch": 0.8389261744966443, |
| "step": 250 |
| }, |
| { |
| "loss": 0.2201, |
| "grad_norm": 0.19683024287223816, |
| "learning_rate": 9.361827222861241e-05, |
| "entropy": 0.22703441381454467, |
| "num_tokens": 4071117.0, |
| "mean_token_accuracy": 0.9421917259693146, |
| "epoch": 0.87248322147651, |
| "step": 260 |
| }, |
| { |
| "loss": 0.2048, |
| "grad_norm": 0.184284046292305, |
| "learning_rate": 9.308597683653975e-05, |
| "entropy": 0.20578452795743943, |
| "num_tokens": 4226750.0, |
| "mean_token_accuracy": 0.9451743960380554, |
| "epoch": 0.9060402684563759, |
| "step": 270 |
| }, |
| { |
| "loss": 0.2181, |
| "grad_norm": 0.19832965731620789, |
| "learning_rate": 9.253400328436699e-05, |
| "entropy": 0.22092150747776032, |
| "num_tokens": 4385096.0, |
| "mean_token_accuracy": 0.9426127344369888, |
| "epoch": 0.9395973154362416, |
| "step": 280 |
| }, |
| { |
| "loss": 0.2178, |
| "grad_norm": 0.16237328946590424, |
| "learning_rate": 9.196260366859342e-05, |
| "entropy": 0.22047539427876472, |
| "num_tokens": 4544552.0, |
| "mean_token_accuracy": 0.9419336885213851, |
| "epoch": 0.9731543624161074, |
| "step": 290 |
| }, |
| { |
| "loss": 0.2188, |
| "grad_norm": 0.19606098532676697, |
| "learning_rate": 9.137203895795983e-05, |
| "entropy": 0.2207832932472229, |
| "num_tokens": 4701426.0, |
| "mean_token_accuracy": 0.941154745221138, |
| "epoch": 1.0067114093959733, |
| "step": 300 |
| }, |
| { |
| "loss": 0.1948, |
| "grad_norm": 0.2281929850578308, |
| "learning_rate": 9.076257887425923e-05, |
| "entropy": 0.19738901741802692, |
| "num_tokens": 4858846.0, |
| "mean_token_accuracy": 0.9471817642450333, |
| "epoch": 1.0402684563758389, |
| "step": 310 |
| }, |
| { |
| "loss": 0.2037, |
| "grad_norm": 0.19096426665782928, |
| "learning_rate": 9.01345017691499e-05, |
| "entropy": 0.2047410562634468, |
| "num_tokens": 5015590.0, |
| "mean_token_accuracy": 0.9454032570123673, |
| "epoch": 1.0738255033557047, |
| "step": 320 |
| }, |
| { |
| "loss": 0.2023, |
| "grad_norm": 0.19078002870082855, |
| "learning_rate": 8.948809449702711e-05, |
| "entropy": 0.21062878221273423, |
| "num_tokens": 5173149.0, |
| "mean_token_accuracy": 0.9446896910667419, |
| "epoch": 1.1073825503355705, |
| "step": 330 |
| }, |
| { |
| "loss": 0.1997, |
| "grad_norm": 0.26884037256240845, |
| "learning_rate": 8.882365228401139e-05, |
| "entropy": 0.20285834297537803, |
| "num_tokens": 5331307.0, |
| "mean_token_accuracy": 0.9449720442295074, |
| "epoch": 1.1409395973154361, |
| "step": 340 |
| }, |
| { |
| "loss": 0.1967, |
| "grad_norm": 0.23782381415367126, |
| "learning_rate": 8.814147859311332e-05, |
| "entropy": 0.1966269500553608, |
| "num_tokens": 5486558.0, |
| "mean_token_accuracy": 0.9466675192117691, |
| "epoch": 1.174496644295302, |
| "step": 350 |
| }, |
| { |
| "loss": 0.2135, |
| "grad_norm": 0.19140051305294037, |
| "learning_rate": 8.744188498563641e-05, |
| "entropy": 0.21856041625142097, |
| "num_tokens": 5643311.0, |
| "mean_token_accuracy": 0.9413786977529526, |
| "epoch": 1.2080536912751678, |
| "step": 360 |
| }, |
| { |
| "loss": 0.1956, |
| "grad_norm": 0.2523305118083954, |
| "learning_rate": 8.672519097888126e-05, |
| "entropy": 0.19831475540995597, |
| "num_tokens": 5798511.0, |
| "mean_token_accuracy": 0.946366423368454, |
| "epoch": 1.2416107382550337, |
| "step": 370 |
| }, |
| { |
| "loss": 0.1895, |
| "grad_norm": 0.16881771385669708, |
| "learning_rate": 8.599172390021615e-05, |
| "entropy": 0.1935427539050579, |
| "num_tokens": 5954868.0, |
| "mean_token_accuracy": 0.9482783049345016, |
| "epoch": 1.2751677852348993, |
| "step": 380 |
| }, |
| { |
| "loss": 0.1857, |
| "grad_norm": 0.15226209163665771, |
| "learning_rate": 8.524181873758059e-05, |
| "entropy": 0.18582973293960095, |
| "num_tokens": 6108894.0, |
| "mean_token_accuracy": 0.9488643199205399, |
| "epoch": 1.308724832214765, |
| "step": 390 |
| }, |
| { |
| "loss": 0.2078, |
| "grad_norm": 0.19812646508216858, |
| "learning_rate": 8.447581798649014e-05, |
| "entropy": 0.20552278831601142, |
| "num_tokens": 6267764.0, |
| "mean_token_accuracy": 0.9442088425159454, |
| "epoch": 1.342281879194631, |
| "step": 400 |
| }, |
| { |
| "loss": 0.2074, |
| "grad_norm": 0.188148632645607, |
| "learning_rate": 8.369407149361241e-05, |
| "entropy": 0.21127415373921393, |
| "num_tokens": 6424716.0, |
| "mean_token_accuracy": 0.9449205726385117, |
| "epoch": 1.3758389261744965, |
| "step": 410 |
| }, |
| { |
| "loss": 0.1948, |
| "grad_norm": 0.19974718987941742, |
| "learning_rate": 8.289693629698564e-05, |
| "entropy": 0.19758303910493852, |
| "num_tokens": 6579760.0, |
| "mean_token_accuracy": 0.9469632744789124, |
| "epoch": 1.4093959731543624, |
| "step": 420 |
| }, |
| { |
| "loss": 0.2144, |
| "grad_norm": 0.2088315635919571, |
| "learning_rate": 8.208477646295277e-05, |
| "entropy": 0.21340604722499848, |
| "num_tokens": 6737911.0, |
| "mean_token_accuracy": 0.9427065312862396, |
| "epoch": 1.4429530201342282, |
| "step": 430 |
| }, |
| { |
| "loss": 0.2145, |
| "grad_norm": 0.31156837940216064, |
| "learning_rate": 8.125796291988577e-05, |
| "entropy": 0.21349589377641678, |
| "num_tokens": 6894903.0, |
| "mean_token_accuracy": 0.9425385951995849, |
| "epoch": 1.476510067114094, |
| "step": 440 |
| }, |
| { |
| "loss": 0.2159, |
| "grad_norm": 0.18016791343688965, |
| "learning_rate": 8.041687328877567e-05, |
| "entropy": 0.2204777255654335, |
| "num_tokens": 7054943.0, |
| "mean_token_accuracy": 0.9415547668933868, |
| "epoch": 1.5100671140939599, |
| "step": 450 |
| }, |
| { |
| "loss": 0.202, |
| "grad_norm": 0.2058987319469452, |
| "learning_rate": 7.956189171076616e-05, |
| "entropy": 0.20117317102849483, |
| "num_tokens": 7211170.0, |
| "mean_token_accuracy": 0.9456082373857498, |
| "epoch": 1.5436241610738255, |
| "step": 460 |
| }, |
| { |
| "loss": 0.1939, |
| "grad_norm": 0.15302836894989014, |
| "learning_rate": 7.869340867170928e-05, |
| "entropy": 0.19479563012719153, |
| "num_tokens": 7369121.0, |
| "mean_token_accuracy": 0.9471449553966522, |
| "epoch": 1.5771812080536913, |
| "step": 470 |
| }, |
| { |
| "loss": 0.1886, |
| "grad_norm": 0.16370812058448792, |
| "learning_rate": 7.781182082382325e-05, |
| "entropy": 0.18975077904760837, |
| "num_tokens": 7525693.0, |
| "mean_token_accuracy": 0.9485057532787323, |
| "epoch": 1.610738255033557, |
| "step": 480 |
| }, |
| { |
| "loss": 0.2023, |
| "grad_norm": 0.1960698664188385, |
| "learning_rate": 7.691753080453412e-05, |
| "entropy": 0.2045104220509529, |
| "num_tokens": 7683086.0, |
| "mean_token_accuracy": 0.9456699937582016, |
| "epoch": 1.6442953020134228, |
| "step": 490 |
| }, |
| { |
| "loss": 0.1991, |
| "grad_norm": 0.18973101675510406, |
| "learning_rate": 7.60109470525839e-05, |
| "entropy": 0.20317464768886567, |
| "num_tokens": 7840651.0, |
| "mean_token_accuracy": 0.9453895062208175, |
| "epoch": 1.6778523489932886, |
| "step": 500 |
| }, |
| { |
| "loss": 0.1939, |
| "grad_norm": 0.19162558019161224, |
| "learning_rate": 7.509248362148889e-05, |
| "entropy": 0.19502840861678122, |
| "num_tokens": 7997627.0, |
| "mean_token_accuracy": 0.9481672704219818, |
| "epoch": 1.7114093959731544, |
| "step": 510 |
| }, |
| { |
| "loss": 0.1921, |
| "grad_norm": 0.18042920529842377, |
| "learning_rate": 7.416255999043401e-05, |
| "entropy": 0.19183044098317623, |
| "num_tokens": 8153075.0, |
| "mean_token_accuracy": 0.9467999368906022, |
| "epoch": 1.7449664429530203, |
| "step": 520 |
| }, |
| { |
| "loss": 0.1788, |
| "grad_norm": 0.1608305722475052, |
| "learning_rate": 7.322160087268877e-05, |
| "entropy": 0.18251866959035395, |
| "num_tokens": 8307904.0, |
| "mean_token_accuracy": 0.9506349295377732, |
| "epoch": 1.778523489932886, |
| "step": 530 |
| }, |
| { |
| "loss": 0.1808, |
| "grad_norm": 0.18992868065834045, |
| "learning_rate": 7.227003602163295e-05, |
| "entropy": 0.1834562122821808, |
| "num_tokens": 8464236.0, |
| "mean_token_accuracy": 0.9505033463239669, |
| "epoch": 1.8120805369127517, |
| "step": 540 |
| }, |
| { |
| "loss": 0.1868, |
| "grad_norm": 0.21314968168735504, |
| "learning_rate": 7.130830003448032e-05, |
| "entropy": 0.1886416744440794, |
| "num_tokens": 8622442.0, |
| "mean_token_accuracy": 0.9492894530296325, |
| "epoch": 1.8456375838926173, |
| "step": 550 |
| }, |
| { |
| "loss": 0.19, |
| "grad_norm": 0.196988046169281, |
| "learning_rate": 7.033683215379002e-05, |
| "entropy": 0.1898799568414688, |
| "num_tokens": 8778712.0, |
| "mean_token_accuracy": 0.9487443953752518, |
| "epoch": 1.8791946308724832, |
| "step": 560 |
| }, |
| { |
| "loss": 0.1872, |
| "grad_norm": 0.19480957090854645, |
| "learning_rate": 6.935607606685642e-05, |
| "entropy": 0.1890778660774231, |
| "num_tokens": 8934261.0, |
| "mean_token_accuracy": 0.9492935538291931, |
| "epoch": 1.912751677852349, |
| "step": 570 |
| }, |
| { |
| "loss": 0.1923, |
| "grad_norm": 0.1902640014886856, |
| "learning_rate": 6.836647970306894e-05, |
| "entropy": 0.1931080285459757, |
| "num_tokens": 9093573.0, |
| "mean_token_accuracy": 0.9487350136041641, |
| "epoch": 1.9463087248322148, |
| "step": 580 |
| }, |
| { |
| "loss": 0.1817, |
| "grad_norm": 0.19666995108127594, |
| "learning_rate": 6.736849502933452e-05, |
| "entropy": 0.18354742750525474, |
| "num_tokens": 9247665.0, |
| "mean_token_accuracy": 0.9502300530672073, |
| "epoch": 1.9798657718120807, |
| "step": 590 |
| }, |
| { |
| "loss": 0.1859, |
| "grad_norm": 0.2059757262468338, |
| "learning_rate": 6.636257784365584e-05, |
| "entropy": 0.18999958783388138, |
| "num_tokens": 9402433.0, |
| "mean_token_accuracy": 0.9492688685655594, |
| "epoch": 2.0134228187919465, |
| "step": 600 |
| }, |
| { |
| "loss": 0.1847, |
| "grad_norm": 0.1790536344051361, |
| "learning_rate": 6.53491875669601e-05, |
| "entropy": 0.18768733143806457, |
| "num_tokens": 9559144.0, |
| "mean_token_accuracy": 0.949212983250618, |
| "epoch": 2.046979865771812, |
| "step": 610 |
| }, |
| { |
| "loss": 0.1642, |
| "grad_norm": 0.21339593827724457, |
| "learning_rate": 6.432878703327298e-05, |
| "entropy": 0.16429513394832612, |
| "num_tokens": 9714865.0, |
| "mean_token_accuracy": 0.9541611701250077, |
| "epoch": 2.0805369127516777, |
| "step": 620 |
| }, |
| { |
| "loss": 0.1855, |
| "grad_norm": 0.21740688383579254, |
| "learning_rate": 6.330184227833376e-05, |
| "entropy": 0.18604125529527665, |
| "num_tokens": 9872250.0, |
| "mean_token_accuracy": 0.9479648023843765, |
| "epoch": 2.1140939597315436, |
| "step": 630 |
| }, |
| { |
| "loss": 0.1898, |
| "grad_norm": 0.1908544898033142, |
| "learning_rate": 6.226882232674825e-05, |
| "entropy": 0.19123471677303314, |
| "num_tokens": 10026954.0, |
| "mean_token_accuracy": 0.9476721823215485, |
| "epoch": 2.1476510067114094, |
| "step": 640 |
| }, |
| { |
| "loss": 0.1652, |
| "grad_norm": 0.2003369778394699, |
| "learning_rate": 6.123019897777657e-05, |
| "entropy": 0.16700895316898823, |
| "num_tokens": 10177873.0, |
| "mean_token_accuracy": 0.9538896352052688, |
| "epoch": 2.1812080536912752, |
| "step": 650 |
| }, |
| { |
| "loss": 0.2036, |
| "grad_norm": 0.22211240231990814, |
| "learning_rate": 6.0186446589853784e-05, |
| "entropy": 0.20571780651807786, |
| "num_tokens": 10338481.0, |
| "mean_token_accuracy": 0.9439575403928757, |
| "epoch": 2.214765100671141, |
| "step": 660 |
| }, |
| { |
| "loss": 0.2002, |
| "grad_norm": 0.23921112716197968, |
| "learning_rate": 5.9138041863941616e-05, |
| "entropy": 0.19964271709322928, |
| "num_tokens": 10495720.0, |
| "mean_token_accuracy": 0.9459815502166748, |
| "epoch": 2.248322147651007, |
| "step": 670 |
| }, |
| { |
| "loss": 0.1989, |
| "grad_norm": 0.20520520210266113, |
| "learning_rate": 5.808546362581032e-05, |
| "entropy": 0.20052680410444737, |
| "num_tokens": 10655224.0, |
| "mean_token_accuracy": 0.9459189057350159, |
| "epoch": 2.2818791946308723, |
| "step": 680 |
| }, |
| { |
| "loss": 0.1803, |
| "grad_norm": 0.20859530568122864, |
| "learning_rate": 5.7029192607350146e-05, |
| "entropy": 0.18437107987701892, |
| "num_tokens": 10812393.0, |
| "mean_token_accuracy": 0.9512098997831344, |
| "epoch": 2.315436241610738, |
| "step": 690 |
| }, |
| { |
| "loss": 0.1625, |
| "grad_norm": 0.21114759147167206, |
| "learning_rate": 5.596971122701221e-05, |
| "entropy": 0.16424274519085885, |
| "num_tokens": 10965653.0, |
| "mean_token_accuracy": 0.9543471187353134, |
| "epoch": 2.348993288590604, |
| "step": 700 |
| }, |
| { |
| "loss": 0.164, |
| "grad_norm": 0.19510741531848907, |
| "learning_rate": 5.4907503369479116e-05, |
| "entropy": 0.16679177805781364, |
| "num_tokens": 11120539.0, |
| "mean_token_accuracy": 0.954514691233635, |
| "epoch": 2.38255033557047, |
| "step": 710 |
| }, |
| { |
| "loss": 0.1772, |
| "grad_norm": 0.244350403547287, |
| "learning_rate": 5.384305416466584e-05, |
| "entropy": 0.17797317542135715, |
| "num_tokens": 11275790.0, |
| "mean_token_accuracy": 0.950087907910347, |
| "epoch": 2.4161073825503356, |
| "step": 720 |
| }, |
| { |
| "loss": 0.1835, |
| "grad_norm": 0.19845031201839447, |
| "learning_rate": 5.2776849766152e-05, |
| "entropy": 0.18638909384608268, |
| "num_tokens": 11430204.0, |
| "mean_token_accuracy": 0.9490520358085632, |
| "epoch": 2.4496644295302015, |
| "step": 730 |
| }, |
| { |
| "loss": 0.1883, |
| "grad_norm": 0.20039933919906616, |
| "learning_rate": 5.170937712914655e-05, |
| "entropy": 0.18770763501524926, |
| "num_tokens": 11592345.0, |
| "mean_token_accuracy": 0.9483928889036178, |
| "epoch": 2.4832214765100673, |
| "step": 740 |
| }, |
| { |
| "loss": 0.1528, |
| "grad_norm": 0.20752805471420288, |
| "learning_rate": 5.064112378808637e-05, |
| "entropy": 0.15445428304374217, |
| "num_tokens": 11744570.0, |
| "mean_token_accuracy": 0.9564135998487473, |
| "epoch": 2.5167785234899327, |
| "step": 750 |
| }, |
| { |
| "loss": 0.1793, |
| "grad_norm": 0.2342364341020584, |
| "learning_rate": 4.957257763397024e-05, |
| "entropy": 0.18113909810781478, |
| "num_tokens": 11901062.0, |
| "mean_token_accuracy": 0.9501085758209229, |
| "epoch": 2.5503355704697985, |
| "step": 760 |
| }, |
| { |
| "loss": 0.1572, |
| "grad_norm": 0.20083309710025787, |
| "learning_rate": 4.850422669153009e-05, |
| "entropy": 0.16168876476585864, |
| "num_tokens": 12053753.0, |
| "mean_token_accuracy": 0.9557742238044739, |
| "epoch": 2.5838926174496644, |
| "step": 770 |
| }, |
| { |
| "loss": 0.1625, |
| "grad_norm": 0.20891757309436798, |
| "learning_rate": 4.743655889634105e-05, |
| "entropy": 0.16362526044249534, |
| "num_tokens": 12208995.0, |
| "mean_token_accuracy": 0.9552077889442444, |
| "epoch": 2.61744966442953, |
| "step": 780 |
| }, |
| { |
| "loss": 0.1567, |
| "grad_norm": 0.21302001178264618, |
| "learning_rate": 4.6370061871972326e-05, |
| "entropy": 0.15993254519999028, |
| "num_tokens": 12364530.0, |
| "mean_token_accuracy": 0.955506107211113, |
| "epoch": 2.651006711409396, |
| "step": 790 |
| }, |
| { |
| "loss": 0.1705, |
| "grad_norm": 0.19647525250911713, |
| "learning_rate": 4.530522270728048e-05, |
| "entropy": 0.1731419090181589, |
| "num_tokens": 12522512.0, |
| "mean_token_accuracy": 0.9523056238889694, |
| "epoch": 2.684563758389262, |
| "step": 800 |
| }, |
| { |
| "loss": 0.1656, |
| "grad_norm": 0.23564544320106506, |
| "learning_rate": 4.424252773394704e-05, |
| "entropy": 0.16778925359249114, |
| "num_tokens": 12678022.0, |
| "mean_token_accuracy": 0.9539547711610794, |
| "epoch": 2.7181208053691277, |
| "step": 810 |
| }, |
| { |
| "loss": 0.1862, |
| "grad_norm": 0.19549453258514404, |
| "learning_rate": 4.318246230436174e-05, |
| "entropy": 0.1878517519682646, |
| "num_tokens": 12838063.0, |
| "mean_token_accuracy": 0.949842843413353, |
| "epoch": 2.751677852348993, |
| "step": 820 |
| }, |
| { |
| "loss": 0.1687, |
| "grad_norm": 0.23449555039405823, |
| "learning_rate": 4.212551056995323e-05, |
| "entropy": 0.17150973305106162, |
| "num_tokens": 12993073.0, |
| "mean_token_accuracy": 0.953241491317749, |
| "epoch": 2.785234899328859, |
| "step": 830 |
| }, |
| { |
| "loss": 0.181, |
| "grad_norm": 0.19433574378490448, |
| "learning_rate": 4.107215526006817e-05, |
| "entropy": 0.1791513752192259, |
| "num_tokens": 13151000.0, |
| "mean_token_accuracy": 0.9498830765485764, |
| "epoch": 2.8187919463087248, |
| "step": 840 |
| }, |
| { |
| "loss": 0.1782, |
| "grad_norm": 0.31362298130989075, |
| "learning_rate": 4.00228774614998e-05, |
| "entropy": 0.18104747980833052, |
| "num_tokens": 13306740.0, |
| "mean_token_accuracy": 0.9511455327272416, |
| "epoch": 2.8523489932885906, |
| "step": 850 |
| }, |
| { |
| "loss": 0.1841, |
| "grad_norm": 0.19440777599811554, |
| "learning_rate": 3.897815639876673e-05, |
| "entropy": 0.183981429412961, |
| "num_tokens": 13465413.0, |
| "mean_token_accuracy": 0.9499198466539382, |
| "epoch": 2.8859060402684564, |
| "step": 860 |
| }, |
| { |
| "loss": 0.1727, |
| "grad_norm": 0.22845511138439178, |
| "learning_rate": 3.793846921524237e-05, |
| "entropy": 0.17417334467172624, |
| "num_tokens": 13625300.0, |
| "mean_token_accuracy": 0.9522460520267486, |
| "epoch": 2.9194630872483223, |
| "step": 870 |
| }, |
| { |
| "loss": 0.1917, |
| "grad_norm": 0.17185327410697937, |
| "learning_rate": 3.6904290755234604e-05, |
| "entropy": 0.19201257564127444, |
| "num_tokens": 13787633.0, |
| "mean_token_accuracy": 0.9470277667045593, |
| "epoch": 2.953020134228188, |
| "step": 880 |
| }, |
| { |
| "loss": 0.181, |
| "grad_norm": 0.22177894413471222, |
| "learning_rate": 3.587609334711576e-05, |
| "entropy": 0.18371602855622768, |
| "num_tokens": 13946986.0, |
| "mean_token_accuracy": 0.9502452671527862, |
| "epoch": 2.9865771812080535, |
| "step": 890 |
| }, |
| { |
| "loss": 0.1639, |
| "grad_norm": 0.20057617127895355, |
| "learning_rate": 3.48543465876014e-05, |
| "entropy": 0.16733385995030403, |
| "num_tokens": 14102933.0, |
| "mean_token_accuracy": 0.9548174262046814, |
| "epoch": 3.0201342281879193, |
| "step": 900 |
| }, |
| { |
| "loss": 0.1657, |
| "grad_norm": 0.234299898147583, |
| "learning_rate": 3.383951712727701e-05, |
| "entropy": 0.1675445105880499, |
| "num_tokens": 14259753.0, |
| "mean_token_accuracy": 0.9535206586122513, |
| "epoch": 3.053691275167785, |
| "step": 910 |
| }, |
| { |
| "loss": 0.1583, |
| "grad_norm": 0.23432078957557678, |
| "learning_rate": 3.2832068457469945e-05, |
| "entropy": 0.1605758022516966, |
| "num_tokens": 14418531.0, |
| "mean_token_accuracy": 0.955757224559784, |
| "epoch": 3.087248322147651, |
| "step": 920 |
| }, |
| { |
| "loss": 0.1504, |
| "grad_norm": 0.20643459260463715, |
| "learning_rate": 3.183246069856443e-05, |
| "entropy": 0.15479738190770148, |
| "num_tokens": 14573081.0, |
| "mean_token_accuracy": 0.9577652394771576, |
| "epoch": 3.120805369127517, |
| "step": 930 |
| }, |
| { |
| "loss": 0.1616, |
| "grad_norm": 0.2060076892375946, |
| "learning_rate": 3.0841150389856125e-05, |
| "entropy": 0.16488183476030827, |
| "num_tokens": 14727805.0, |
| "mean_token_accuracy": 0.9542657196521759, |
| "epoch": 3.1543624161073827, |
| "step": 940 |
| }, |
| { |
| "loss": 0.1727, |
| "grad_norm": 0.25925537943840027, |
| "learning_rate": 2.9858590281042152e-05, |
| "entropy": 0.17419612146914004, |
| "num_tokens": 14883574.0, |
| "mean_token_accuracy": 0.9522079944610595, |
| "epoch": 3.1879194630872485, |
| "step": 950 |
| }, |
| { |
| "loss": 0.1444, |
| "grad_norm": 0.22550174593925476, |
| "learning_rate": 2.888522912544202e-05, |
| "entropy": 0.1470055654644966, |
| "num_tokens": 15038732.0, |
| "mean_token_accuracy": 0.9591749131679534, |
| "epoch": 3.221476510067114, |
| "step": 960 |
| }, |
| { |
| "loss": 0.1453, |
| "grad_norm": 0.22862796485424042, |
| "learning_rate": 2.792151147504366e-05, |
| "entropy": 0.14832949005067347, |
| "num_tokens": 15193067.0, |
| "mean_token_accuracy": 0.9592784523963929, |
| "epoch": 3.2550335570469797, |
| "step": 970 |
| }, |
| { |
| "loss": 0.1592, |
| "grad_norm": 0.21343277394771576, |
| "learning_rate": 2.6967877477468397e-05, |
| "entropy": 0.16025195196270942, |
| "num_tokens": 15348668.0, |
| "mean_token_accuracy": 0.9546813875436783, |
| "epoch": 3.2885906040268456, |
| "step": 980 |
| }, |
| { |
| "loss": 0.1842, |
| "grad_norm": 0.27152112126350403, |
| "learning_rate": 2.6024762674947313e-05, |
| "entropy": 0.18718509674072265, |
| "num_tokens": 15507728.0, |
| "mean_token_accuracy": 0.9490089803934098, |
| "epoch": 3.3221476510067114, |
| "step": 990 |
| }, |
| { |
| "loss": 0.1613, |
| "grad_norm": 0.20685099065303802, |
| "learning_rate": 2.509259780540118e-05, |
| "entropy": 0.16506212055683137, |
| "num_tokens": 15663068.0, |
| "mean_token_accuracy": 0.9536582648754119, |
| "epoch": 3.3557046979865772, |
| "step": 1000 |
| }, |
| { |
| "loss": 0.1557, |
| "grad_norm": 0.24159930646419525, |
| "learning_rate": 2.4171808605714504e-05, |
| "entropy": 0.15720976293087005, |
| "num_tokens": 15820651.0, |
| "mean_token_accuracy": 0.9559206962585449, |
| "epoch": 3.389261744966443, |
| "step": 1010 |
| }, |
| { |
| "loss": 0.1775, |
| "grad_norm": 0.23764103651046753, |
| "learning_rate": 2.3262815617293517e-05, |
| "entropy": 0.1807922873646021, |
| "num_tokens": 15979358.0, |
| "mean_token_accuracy": 0.9511760205030442, |
| "epoch": 3.422818791946309, |
| "step": 1020 |
| }, |
| { |
| "loss": 0.1713, |
| "grad_norm": 0.22108638286590576, |
| "learning_rate": 2.2366033993997344e-05, |
| "entropy": 0.17095785699784755, |
| "num_tokens": 16135558.0, |
| "mean_token_accuracy": 0.95192691385746, |
| "epoch": 3.4563758389261743, |
| "step": 1030 |
| }, |
| { |
| "loss": 0.1682, |
| "grad_norm": 0.24025718867778778, |
| "learning_rate": 2.1481873312529426e-05, |
| "entropy": 0.17155840508639814, |
| "num_tokens": 16289493.0, |
| "mean_token_accuracy": 0.9537844061851501, |
| "epoch": 3.48993288590604, |
| "step": 1040 |
| }, |
| { |
| "loss": 0.1645, |
| "grad_norm": 0.22691352665424347, |
| "learning_rate": 2.061073738537635e-05, |
| "entropy": 0.1673688843846321, |
| "num_tokens": 16444253.0, |
| "mean_token_accuracy": 0.9540765851736068, |
| "epoch": 3.523489932885906, |
| "step": 1050 |
| }, |
| { |
| "loss": 0.1711, |
| "grad_norm": 0.260170042514801, |
| "learning_rate": 1.975302407637929e-05, |
| "entropy": 0.17190844528377056, |
| "num_tokens": 16605899.0, |
| "mean_token_accuracy": 0.9533313482999801, |
| "epoch": 3.557046979865772, |
| "step": 1060 |
| }, |
| { |
| "loss": 0.1703, |
| "grad_norm": 0.22882592678070068, |
| "learning_rate": 1.8909125119022136e-05, |
| "entropy": 0.17202796116471292, |
| "num_tokens": 16762980.0, |
| "mean_token_accuracy": 0.9525971114635468, |
| "epoch": 3.5906040268456376, |
| "step": 1070 |
| }, |
| { |
| "loss": 0.1751, |
| "grad_norm": 0.2285274714231491, |
| "learning_rate": 1.807942593751973e-05, |
| "entropy": 0.17774682715535164, |
| "num_tokens": 16923345.0, |
| "mean_token_accuracy": 0.9512902319431304, |
| "epoch": 3.6241610738255035, |
| "step": 1080 |
| }, |
| { |
| "loss": 0.1577, |
| "grad_norm": 0.24551482498645782, |
| "learning_rate": 1.7264305470787363e-05, |
| "entropy": 0.1609206810593605, |
| "num_tokens": 17077331.0, |
| "mean_token_accuracy": 0.955593678355217, |
| "epoch": 3.6577181208053693, |
| "step": 1090 |
| }, |
| { |
| "loss": 0.1558, |
| "grad_norm": 0.22661259770393372, |
| "learning_rate": 1.6464135999372537e-05, |
| "entropy": 0.1576618704944849, |
| "num_tokens": 17231136.0, |
| "mean_token_accuracy": 0.9559727132320404, |
| "epoch": 3.6912751677852347, |
| "step": 1100 |
| }, |
| { |
| "loss": 0.187, |
| "grad_norm": 0.23124344646930695, |
| "learning_rate": 1.567928297542749e-05, |
| "entropy": 0.18841546326875686, |
| "num_tokens": 17390871.0, |
| "mean_token_accuracy": 0.9487693756818771, |
| "epoch": 3.7248322147651005, |
| "step": 1110 |
| }, |
| { |
| "loss": 0.1368, |
| "grad_norm": 0.2734222710132599, |
| "learning_rate": 1.4910104855800427e-05, |
| "entropy": 0.14009847678244114, |
| "num_tokens": 17544815.0, |
| "mean_token_accuracy": 0.9597841829061509, |
| "epoch": 3.7583892617449663, |
| "step": 1120 |
| }, |
| { |
| "loss": 0.1557, |
| "grad_norm": 0.25066235661506653, |
| "learning_rate": 1.4156952938321798e-05, |
| "entropy": 0.1570753049105406, |
| "num_tokens": 17701337.0, |
| "mean_token_accuracy": 0.9568644523620605, |
| "epoch": 3.791946308724832, |
| "step": 1130 |
| }, |
| { |
| "loss": 0.1582, |
| "grad_norm": 0.24784211814403534, |
| "learning_rate": 1.3420171201359933e-05, |
| "entropy": 0.15978324972093105, |
| "num_tokens": 17855471.0, |
| "mean_token_accuracy": 0.9565310895442962, |
| "epoch": 3.825503355704698, |
| "step": 1140 |
| }, |
| { |
| "loss": 0.1583, |
| "grad_norm": 0.22498124837875366, |
| "learning_rate": 1.2700096146719931e-05, |
| "entropy": 0.16122703738510608, |
| "num_tokens": 18017269.0, |
| "mean_token_accuracy": 0.955768808722496, |
| "epoch": 3.859060402684564, |
| "step": 1150 |
| }, |
| { |
| "loss": 0.1729, |
| "grad_norm": 0.22251562774181366, |
| "learning_rate": 1.1997056645956967e-05, |
| "entropy": 0.17401745431125165, |
| "num_tokens": 18176540.0, |
| "mean_token_accuracy": 0.9513998061418534, |
| "epoch": 3.8926174496644297, |
| "step": 1160 |
| }, |
| { |
| "loss": 0.1429, |
| "grad_norm": 0.2645820677280426, |
| "learning_rate": 1.1311373790174657e-05, |
| "entropy": 0.14712858721613883, |
| "num_tokens": 18331179.0, |
| "mean_token_accuracy": 0.9599408686161042, |
| "epoch": 3.926174496644295, |
| "step": 1170 |
| }, |
| { |
| "loss": 0.1778, |
| "grad_norm": 0.22513671219348907, |
| "learning_rate": 1.0643360743376829e-05, |
| "entropy": 0.178417731449008, |
| "num_tokens": 18491744.0, |
| "mean_token_accuracy": 0.9512795448303223, |
| "epoch": 3.959731543624161, |
| "step": 1180 |
| }, |
| { |
| "loss": 0.1751, |
| "grad_norm": 0.23683220148086548, |
| "learning_rate": 9.993322599439692e-06, |
| "entropy": 0.1774978566914797, |
| "num_tokens": 18647922.0, |
| "mean_token_accuracy": 0.9514397591352463, |
| "epoch": 3.9932885906040267, |
| "step": 1190 |
| }, |
| { |
| "loss": 0.1343, |
| "grad_norm": 0.23946121335029602, |
| "learning_rate": 9.36155624276987e-06, |
| "entropy": 0.13949108868837357, |
| "num_tokens": 18799741.0, |
| "mean_token_accuracy": 0.961316853761673, |
| "epoch": 4.026845637583893, |
| "step": 1200 |
| }, |
| { |
| "loss": 0.1448, |
| "grad_norm": 0.26739007234573364, |
| "learning_rate": 8.748350212711853e-06, |
| "entropy": 0.14928585700690747, |
| "num_tokens": 18954239.0, |
| "mean_token_accuracy": 0.9586472451686859, |
| "epoch": 4.060402684563758, |
| "step": 1210 |
| }, |
| { |
| "loss": 0.1589, |
| "grad_norm": 0.2560782730579376, |
| "learning_rate": 8.153984571766699e-06, |
| "entropy": 0.1610208224505186, |
| "num_tokens": 19113299.0, |
| "mean_token_accuracy": 0.955417987704277, |
| "epoch": 4.093959731543624, |
| "step": 1220 |
| }, |
| { |
| "loss": 0.1566, |
| "grad_norm": 0.2503839135169983, |
| "learning_rate": 7.578730777682386e-06, |
| "entropy": 0.15893662311136722, |
| "num_tokens": 19269704.0, |
| "mean_token_accuracy": 0.9564489126205444, |
| "epoch": 4.12751677852349, |
| "step": 1230 |
| }, |
| { |
| "loss": 0.1729, |
| "grad_norm": 0.2269158661365509, |
| "learning_rate": 7.022851559473964e-06, |
| "entropy": 0.17561149559915065, |
| "num_tokens": 19430349.0, |
| "mean_token_accuracy": 0.95240318775177, |
| "epoch": 4.1610738255033555, |
| "step": 1240 |
| }, |
| { |
| "loss": 0.168, |
| "grad_norm": 0.2869897186756134, |
| "learning_rate": 6.486600797430492e-06, |
| "entropy": 0.1709251943975687, |
| "num_tokens": 19585660.0, |
| "mean_token_accuracy": 0.9534720987081527, |
| "epoch": 4.194630872483222, |
| "step": 1250 |
| }, |
| { |
| "loss": 0.1457, |
| "grad_norm": 0.25000014901161194, |
| "learning_rate": 5.9702234071631e-06, |
| "entropy": 0.14829469434916973, |
| "num_tokens": 19744182.0, |
| "mean_token_accuracy": 0.9593781590461731, |
| "epoch": 4.228187919463087, |
| "step": 1260 |
| }, |
| { |
| "loss": 0.16, |
| "grad_norm": 0.24967129528522491, |
| "learning_rate": 5.473955227747623e-06, |
| "entropy": 0.16518744193017482, |
| "num_tokens": 19902824.0, |
| "mean_token_accuracy": 0.9548085004091262, |
| "epoch": 4.261744966442953, |
| "step": 1270 |
| }, |
| { |
| "loss": 0.1508, |
| "grad_norm": 0.27587053179740906, |
| "learning_rate": 4.9980229140124905e-06, |
| "entropy": 0.15380522198975086, |
| "num_tokens": 20057944.0, |
| "mean_token_accuracy": 0.9581666350364685, |
| "epoch": 4.295302013422819, |
| "step": 1280 |
| }, |
| { |
| "loss": 0.1497, |
| "grad_norm": 0.2772783935070038, |
| "learning_rate": 4.542643833021254e-06, |
| "entropy": 0.15263673029839991, |
| "num_tokens": 20215072.0, |
| "mean_token_accuracy": 0.9581062227487565, |
| "epoch": 4.328859060402684, |
| "step": 1290 |
| }, |
| { |
| "loss": 0.1313, |
| "grad_norm": 0.25381705164909363, |
| "learning_rate": 4.108025964797135e-06, |
| "entropy": 0.13392407521605493, |
| "num_tokens": 20371415.0, |
| "mean_token_accuracy": 0.9618801653385163, |
| "epoch": 4.3624161073825505, |
| "step": 1300 |
| }, |
| { |
| "loss": 0.1442, |
| "grad_norm": 0.318499892950058, |
| "learning_rate": 3.69436780733462e-06, |
| "entropy": 0.1467492014169693, |
| "num_tokens": 20525796.0, |
| "mean_token_accuracy": 0.9589717149734497, |
| "epoch": 4.395973154362416, |
| "step": 1310 |
| }, |
| { |
| "loss": 0.164, |
| "grad_norm": 0.29794761538505554, |
| "learning_rate": 3.3018582859418446e-06, |
| "entropy": 0.16693559624254703, |
| "num_tokens": 20683012.0, |
| "mean_token_accuracy": 0.9544567495584488, |
| "epoch": 4.429530201342282, |
| "step": 1320 |
| }, |
| { |
| "loss": 0.1585, |
| "grad_norm": 0.26992982625961304, |
| "learning_rate": 2.930676666954846e-06, |
| "entropy": 0.16096150763332845, |
| "num_tokens": 20836880.0, |
| "mean_token_accuracy": 0.955687940120697, |
| "epoch": 4.4630872483221475, |
| "step": 1330 |
| }, |
| { |
| "loss": 0.1545, |
| "grad_norm": 0.2633298635482788, |
| "learning_rate": 2.580992475863381e-06, |
| "entropy": 0.15808030292391778, |
| "num_tokens": 20991520.0, |
| "mean_token_accuracy": 0.9564832031726838, |
| "epoch": 4.496644295302014, |
| "step": 1340 |
| }, |
| { |
| "loss": 0.1574, |
| "grad_norm": 0.2697800099849701, |
| "learning_rate": 2.2529654198854835e-06, |
| "entropy": 0.16020409800112248, |
| "num_tokens": 21149049.0, |
| "mean_token_accuracy": 0.9553818762302398, |
| "epoch": 4.530201342281879, |
| "step": 1350 |
| }, |
| { |
| "loss": 0.1434, |
| "grad_norm": 0.26713675260543823, |
| "learning_rate": 1.9467453150262327e-06, |
| "entropy": 0.1468834660947323, |
| "num_tokens": 21303482.0, |
| "mean_token_accuracy": 0.9584787577390671, |
| "epoch": 4.563758389261745, |
| "step": 1360 |
| }, |
| { |
| "loss": 0.1676, |
| "grad_norm": 0.25141027569770813, |
| "learning_rate": 1.6624720176540265e-06, |
| "entropy": 0.17059922069311143, |
| "num_tokens": 21462214.0, |
| "mean_token_accuracy": 0.9529773116111755, |
| "epoch": 4.597315436241611, |
| "step": 1370 |
| }, |
| { |
| "loss": 0.1631, |
| "grad_norm": 0.22781124711036682, |
| "learning_rate": 1.400275360625608e-06, |
| "entropy": 0.16469677537679672, |
| "num_tokens": 21617768.0, |
| "mean_token_accuracy": 0.9540845274925231, |
| "epoch": 4.630872483221476, |
| "step": 1380 |
| }, |
| { |
| "loss": 0.1405, |
| "grad_norm": 0.26300156116485596, |
| "learning_rate": 1.1602750939889774e-06, |
| "entropy": 0.14349478296935558, |
| "num_tokens": 21775827.0, |
| "mean_token_accuracy": 0.9608874082565307, |
| "epoch": 4.6644295302013425, |
| "step": 1390 |
| }, |
| { |
| "loss": 0.1648, |
| "grad_norm": 0.28018462657928467, |
| "learning_rate": 9.42580830291373e-07, |
| "entropy": 0.168378734216094, |
| "num_tokens": 21933598.0, |
| "mean_token_accuracy": 0.9548741400241851, |
| "epoch": 4.697986577181208, |
| "step": 1400 |
| }, |
| { |
| "loss": 0.1579, |
| "grad_norm": 0.23573631048202515, |
| "learning_rate": 7.472919945171631e-07, |
| "entropy": 0.16024879328906536, |
| "num_tokens": 22092356.0, |
| "mean_token_accuracy": 0.9563170611858368, |
| "epoch": 4.731543624161074, |
| "step": 1410 |
| }, |
| { |
| "loss": 0.1486, |
| "grad_norm": 0.26423606276512146, |
| "learning_rate": 5.74497778678662e-07, |
| "entropy": 0.153883695602417, |
| "num_tokens": 22245245.0, |
| "mean_token_accuracy": 0.9581821829080581, |
| "epoch": 4.76510067114094, |
| "step": 1420 |
| }, |
| { |
| "loss": 0.1668, |
| "grad_norm": 0.24437865614891052, |
| "learning_rate": 4.242771010804558e-07, |
| "entropy": 0.16868923045694828, |
| "num_tokens": 22403545.0, |
| "mean_token_accuracy": 0.9531532049179077, |
| "epoch": 4.798657718120805, |
| "step": 1430 |
| }, |
| { |
| "loss": 0.1497, |
| "grad_norm": 0.2679818868637085, |
| "learning_rate": 2.966985702759828e-07, |
| "entropy": 0.15332721956074238, |
| "num_tokens": 22559658.0, |
| "mean_token_accuracy": 0.9572977095842361, |
| "epoch": 4.832214765100671, |
| "step": 1440 |
| }, |
| { |
| "loss": 0.1487, |
| "grad_norm": 0.25875014066696167, |
| "learning_rate": 1.9182045373273838e-07, |
| "entropy": 0.15049569718539715, |
| "num_tokens": 22718226.0, |
| "mean_token_accuracy": 0.9580355733633041, |
| "epoch": 4.865771812080537, |
| "step": 1450 |
| }, |
| { |
| "loss": 0.1444, |
| "grad_norm": 0.25537535548210144, |
| "learning_rate": 1.0969065122041766e-07, |
| "entropy": 0.14743491858243943, |
| "num_tokens": 22873841.0, |
| "mean_token_accuracy": 0.9597749501466751, |
| "epoch": 4.899328859060403, |
| "step": 1460 |
| }, |
| { |
| "loss": 0.1584, |
| "grad_norm": 0.28133252263069153, |
| "learning_rate": 5.0346672934270534e-08, |
| "entropy": 0.16207893192768097, |
| "num_tokens": 23032944.0, |
| "mean_token_accuracy": 0.9557940989732743, |
| "epoch": 4.932885906040268, |
| "step": 1470 |
| }, |
| { |
| "loss": 0.1577, |
| "grad_norm": 0.25087353587150574, |
| "learning_rate": 1.3815622363427815e-08, |
| "entropy": 0.16037070676684378, |
| "num_tokens": 23191641.0, |
| "mean_token_accuracy": 0.9560932219028473, |
| "epoch": 4.966442953020135, |
| "step": 1480 |
| }, |
| { |
| "loss": 0.1779, |
| "grad_norm": 0.307130366563797, |
| "learning_rate": 1.141839123142141e-10, |
| "entropy": 0.18243111334741116, |
| "num_tokens": 23348865.0, |
| "mean_token_accuracy": 0.9507445961236953, |
| "epoch": 5.0, |
| "step": 1490 |
| }, |
| { |
| "train_runtime": 10055.6246, |
| "train_samples_per_second": 4.74, |
| "train_steps_per_second": 0.148, |
| "total_flos": 1.0796462610038784e+19, |
| "train_loss": 0.2056295836531876, |
| "epoch": 5.0, |
| "step": 1490 |
| } |
| ] |
