| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.7531847133757962, |
| "eval_steps": 500, |
| "global_step": 473, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.007961783439490446, |
| "grad_norm": 7.592441082000732, |
| "learning_rate": 6.53061224489796e-06, |
| "loss": 0.6416, |
| "mean_token_accuracy": 0.8507869958877563, |
| "num_tokens": 339096.0, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.01592356687898089, |
| "grad_norm": 2.8016273975372314, |
| "learning_rate": 1.4693877551020408e-05, |
| "loss": 0.3207, |
| "mean_token_accuracy": 0.9229801297187805, |
| "num_tokens": 674151.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.02388535031847134, |
| "grad_norm": 1.659272313117981, |
| "learning_rate": 2.285714285714286e-05, |
| "loss": 0.2971, |
| "mean_token_accuracy": 0.9176519632339477, |
| "num_tokens": 1012703.0, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.03184713375796178, |
| "grad_norm": 1.890880823135376, |
| "learning_rate": 3.102040816326531e-05, |
| "loss": 0.2835, |
| "mean_token_accuracy": 0.9199264526367188, |
| "num_tokens": 1345716.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.03980891719745223, |
| "grad_norm": 1.7328331470489502, |
| "learning_rate": 3.918367346938776e-05, |
| "loss": 0.279, |
| "mean_token_accuracy": 0.926113897562027, |
| "num_tokens": 1675339.0, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.04777070063694268, |
| "grad_norm": 1.9407950639724731, |
| "learning_rate": 4.734693877551021e-05, |
| "loss": 0.2817, |
| "mean_token_accuracy": 0.9254960477352142, |
| "num_tokens": 2011971.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.05573248407643312, |
| "grad_norm": 1.0245087146759033, |
| "learning_rate": 5.551020408163266e-05, |
| "loss": 0.2516, |
| "mean_token_accuracy": 0.9324373364448547, |
| "num_tokens": 2350036.0, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.06369426751592357, |
| "grad_norm": 1.2446362972259521, |
| "learning_rate": 5.7141722268481806e-05, |
| "loss": 0.2841, |
| "mean_token_accuracy": 0.9243828415870666, |
| "num_tokens": 2690348.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.07165605095541401, |
| "grad_norm": 1.0921810865402222, |
| "learning_rate": 5.713711204735125e-05, |
| "loss": 0.2561, |
| "mean_token_accuracy": 0.933571070432663, |
| "num_tokens": 3024859.0, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.07961783439490445, |
| "grad_norm": 1.595969319343567, |
| "learning_rate": 5.712895631222946e-05, |
| "loss": 0.2435, |
| "mean_token_accuracy": 0.9315800428390503, |
| "num_tokens": 3362755.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.0875796178343949, |
| "grad_norm": 1.3464078903198242, |
| "learning_rate": 5.71172564129044e-05, |
| "loss": 0.2655, |
| "mean_token_accuracy": 0.922149783372879, |
| "num_tokens": 3693448.0, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.09554140127388536, |
| "grad_norm": 0.8314092755317688, |
| "learning_rate": 5.7102014285729185e-05, |
| "loss": 0.2333, |
| "mean_token_accuracy": 0.9335690081119538, |
| "num_tokens": 4027630.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.1035031847133758, |
| "grad_norm": 1.1375393867492676, |
| "learning_rate": 5.708323245330166e-05, |
| "loss": 0.2128, |
| "mean_token_accuracy": 0.937630420923233, |
| "num_tokens": 4367430.0, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.11146496815286625, |
| "grad_norm": 1.3092705011367798, |
| "learning_rate": 5.706091402404683e-05, |
| "loss": 0.2936, |
| "mean_token_accuracy": 0.9235719919204712, |
| "num_tokens": 4703258.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.11942675159235669, |
| "grad_norm": 1.067455768585205, |
| "learning_rate": 5.7035062691702516e-05, |
| "loss": 0.2395, |
| "mean_token_accuracy": 0.9338732600212097, |
| "num_tokens": 5033421.0, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.12738853503184713, |
| "grad_norm": 0.8332642316818237, |
| "learning_rate": 5.700568273470793e-05, |
| "loss": 0.2357, |
| "mean_token_accuracy": 0.9337397754192353, |
| "num_tokens": 5370235.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.13535031847133758, |
| "grad_norm": 0.7727829217910767, |
| "learning_rate": 5.6972779015495635e-05, |
| "loss": 0.3303, |
| "mean_token_accuracy": 0.913148844242096, |
| "num_tokens": 5696363.0, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.14331210191082802, |
| "grad_norm": 1.3631492853164673, |
| "learning_rate": 5.693635697968683e-05, |
| "loss": 0.2123, |
| "mean_token_accuracy": 0.9384756445884704, |
| "num_tokens": 6030294.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.15127388535031847, |
| "grad_norm": 0.7698091268539429, |
| "learning_rate": 5.689642265519002e-05, |
| "loss": 0.2518, |
| "mean_token_accuracy": 0.9307581007480621, |
| "num_tokens": 6360783.0, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.1592356687898089, |
| "grad_norm": 1.0706422328948975, |
| "learning_rate": 5.685298265120349e-05, |
| "loss": 0.246, |
| "mean_token_accuracy": 0.9332472443580627, |
| "num_tokens": 6695223.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16719745222929935, |
| "grad_norm": 1.2837363481521606, |
| "learning_rate": 5.680604415712133e-05, |
| "loss": 0.248, |
| "mean_token_accuracy": 0.9318027794361115, |
| "num_tokens": 7031647.0, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.1751592356687898, |
| "grad_norm": 0.939667820930481, |
| "learning_rate": 5.6755614941343724e-05, |
| "loss": 0.2042, |
| "mean_token_accuracy": 0.9434775173664093, |
| "num_tokens": 7365011.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.18312101910828024, |
| "grad_norm": 0.8467428088188171, |
| "learning_rate": 5.670170334999116e-05, |
| "loss": 0.2779, |
| "mean_token_accuracy": 0.9302612602710724, |
| "num_tokens": 7696325.0, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.1910828025477707, |
| "grad_norm": 0.9731951951980591, |
| "learning_rate": 5.664431830552321e-05, |
| "loss": 0.2095, |
| "mean_token_accuracy": 0.9396341800689697, |
| "num_tokens": 8033371.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.19904458598726116, |
| "grad_norm": 1.043177604675293, |
| "learning_rate": 5.658346930526177e-05, |
| "loss": 0.2427, |
| "mean_token_accuracy": 0.9356390535831451, |
| "num_tokens": 8371585.0, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.2070063694267516, |
| "grad_norm": 0.7214773893356323, |
| "learning_rate": 5.651916641981931e-05, |
| "loss": 0.2101, |
| "mean_token_accuracy": 0.9411853373050689, |
| "num_tokens": 8705164.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.21496815286624205, |
| "grad_norm": 1.2858519554138184, |
| "learning_rate": 5.645142029143213e-05, |
| "loss": 0.2992, |
| "mean_token_accuracy": 0.9250331223011017, |
| "num_tokens": 9035142.0, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.2229299363057325, |
| "grad_norm": 1.1134355068206787, |
| "learning_rate": 5.638024213219905e-05, |
| "loss": 0.2553, |
| "mean_token_accuracy": 0.931336236000061, |
| "num_tokens": 9369010.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.23089171974522293, |
| "grad_norm": 1.180100440979004, |
| "learning_rate": 5.630564372222581e-05, |
| "loss": 0.2568, |
| "mean_token_accuracy": 0.9334300696849823, |
| "num_tokens": 9707439.0, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.23885350318471338, |
| "grad_norm": 1.014606237411499, |
| "learning_rate": 5.622763740767544e-05, |
| "loss": 0.248, |
| "mean_token_accuracy": 0.9289291560649872, |
| "num_tokens": 10037131.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.24681528662420382, |
| "grad_norm": 1.3234199285507202, |
| "learning_rate": 5.614623609872492e-05, |
| "loss": 0.2654, |
| "mean_token_accuracy": 0.9307288646697998, |
| "num_tokens": 10369358.0, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.25477707006369427, |
| "grad_norm": 1.0750676393508911, |
| "learning_rate": 5.606145326742858e-05, |
| "loss": 0.2402, |
| "mean_token_accuracy": 0.9321265935897827, |
| "num_tokens": 10706495.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.2627388535031847, |
| "grad_norm": 1.0729336738586426, |
| "learning_rate": 5.597330294548838e-05, |
| "loss": 0.2505, |
| "mean_token_accuracy": 0.9285952031612397, |
| "num_tokens": 11036376.0, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.27070063694267515, |
| "grad_norm": 1.0413671731948853, |
| "learning_rate": 5.588179972193168e-05, |
| "loss": 0.2818, |
| "mean_token_accuracy": 0.9262890696525574, |
| "num_tokens": 11371676.0, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.2786624203821656, |
| "grad_norm": 0.47623124718666077, |
| "learning_rate": 5.578695874069676e-05, |
| "loss": 0.2064, |
| "mean_token_accuracy": 0.9406647861003876, |
| "num_tokens": 11708617.0, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.28662420382165604, |
| "grad_norm": 0.9347270131111145, |
| "learning_rate": 5.568879569812642e-05, |
| "loss": 0.1891, |
| "mean_token_accuracy": 0.9463615417480469, |
| "num_tokens": 12044297.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.2945859872611465, |
| "grad_norm": 0.8606042265892029, |
| "learning_rate": 5.558732684037023e-05, |
| "loss": 0.3089, |
| "mean_token_accuracy": 0.9213819921016693, |
| "num_tokens": 12376800.0, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.30254777070063693, |
| "grad_norm": 0.6761439442634583, |
| "learning_rate": 5.5482568960695765e-05, |
| "loss": 0.268, |
| "mean_token_accuracy": 0.9293226420879364, |
| "num_tokens": 12713813.0, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.3105095541401274, |
| "grad_norm": 1.0171080827713013, |
| "learning_rate": 5.537453939670929e-05, |
| "loss": 0.2762, |
| "mean_token_accuracy": 0.9235360741615295, |
| "num_tokens": 13041913.0, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.3184713375796178, |
| "grad_norm": 1.0361348390579224, |
| "learning_rate": 5.526325602748635e-05, |
| "loss": 0.2384, |
| "mean_token_accuracy": 0.9319276750087738, |
| "num_tokens": 13379104.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32643312101910826, |
| "grad_norm": 0.5294153690338135, |
| "learning_rate": 5.5148737270612794e-05, |
| "loss": 0.2579, |
| "mean_token_accuracy": 0.9275310754776, |
| "num_tokens": 13713545.0, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.3343949044585987, |
| "grad_norm": 0.49982768297195435, |
| "learning_rate": 5.503100207913655e-05, |
| "loss": 0.2123, |
| "mean_token_accuracy": 0.9392310321331024, |
| "num_tokens": 14051732.0, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.34235668789808915, |
| "grad_norm": 0.8859318494796753, |
| "learning_rate": 5.4910069938430936e-05, |
| "loss": 0.2116, |
| "mean_token_accuracy": 0.9395874381065369, |
| "num_tokens": 14391834.0, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.3503184713375796, |
| "grad_norm": 0.9655736684799194, |
| "learning_rate": 5.47859608629698e-05, |
| "loss": 0.2174, |
| "mean_token_accuracy": 0.9348036348819733, |
| "num_tokens": 14724037.0, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.35828025477707004, |
| "grad_norm": 1.080808401107788, |
| "learning_rate": 5.465869539301502e-05, |
| "loss": 0.2644, |
| "mean_token_accuracy": 0.9275994002819061, |
| "num_tokens": 15056257.0, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.3662420382165605, |
| "grad_norm": 0.757030725479126, |
| "learning_rate": 5.452829459121712e-05, |
| "loss": 0.2229, |
| "mean_token_accuracy": 0.9409208655357361, |
| "num_tokens": 15393655.0, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.37420382165605093, |
| "grad_norm": 1.0483347177505493, |
| "learning_rate": 5.439478003912935e-05, |
| "loss": 0.2494, |
| "mean_token_accuracy": 0.932054477930069, |
| "num_tokens": 15730052.0, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.3821656050955414, |
| "grad_norm": 0.6794143915176392, |
| "learning_rate": 5.425817383363591e-05, |
| "loss": 0.2139, |
| "mean_token_accuracy": 0.9420378267765045, |
| "num_tokens": 16070553.0, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.39012738853503187, |
| "grad_norm": 0.7795118093490601, |
| "learning_rate": 5.411849858329485e-05, |
| "loss": 0.2722, |
| "mean_token_accuracy": 0.9285059630870819, |
| "num_tokens": 16403976.0, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.3980891719745223, |
| "grad_norm": 0.6438557505607605, |
| "learning_rate": 5.3975777404596336e-05, |
| "loss": 0.2501, |
| "mean_token_accuracy": 0.9362667083740235, |
| "num_tokens": 16738438.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.40605095541401276, |
| "grad_norm": 1.1960865259170532, |
| "learning_rate": 5.3830033918136825e-05, |
| "loss": 0.2625, |
| "mean_token_accuracy": 0.9264649868011474, |
| "num_tokens": 17065993.0, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.4140127388535032, |
| "grad_norm": 0.68548583984375, |
| "learning_rate": 5.3681292244709847e-05, |
| "loss": 0.1844, |
| "mean_token_accuracy": 0.947054409980774, |
| "num_tokens": 17408656.0, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.42197452229299365, |
| "grad_norm": 0.7950796484947205, |
| "learning_rate": 5.35295770013139e-05, |
| "loss": 0.2456, |
| "mean_token_accuracy": 0.9316338002681732, |
| "num_tokens": 17744410.0, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.4299363057324841, |
| "grad_norm": 0.7825786471366882, |
| "learning_rate": 5.337491329707838e-05, |
| "loss": 0.2129, |
| "mean_token_accuracy": 0.9396231532096863, |
| "num_tokens": 18078247.0, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.43789808917197454, |
| "grad_norm": 1.0084128379821777, |
| "learning_rate": 5.321732672910793e-05, |
| "loss": 0.2436, |
| "mean_token_accuracy": 0.936199688911438, |
| "num_tokens": 18413898.0, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.445859872611465, |
| "grad_norm": 1.164597988128662, |
| "learning_rate": 5.3056843378246074e-05, |
| "loss": 0.25, |
| "mean_token_accuracy": 0.9329074323177338, |
| "num_tokens": 18746312.0, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.4538216560509554, |
| "grad_norm": 0.8926853537559509, |
| "learning_rate": 5.289348980475881e-05, |
| "loss": 0.2518, |
| "mean_token_accuracy": 0.9305019199848175, |
| "num_tokens": 19075695.0, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.46178343949044587, |
| "grad_norm": 0.592759907245636, |
| "learning_rate": 5.272729304393883e-05, |
| "loss": 0.2326, |
| "mean_token_accuracy": 0.933732408285141, |
| "num_tokens": 19408409.0, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.4697452229299363, |
| "grad_norm": 0.8731628060340881, |
| "learning_rate": 5.2558280601631165e-05, |
| "loss": 0.2301, |
| "mean_token_accuracy": 0.9356718420982361, |
| "num_tokens": 19743820.0, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.47770700636942676, |
| "grad_norm": 1.0021252632141113, |
| "learning_rate": 5.238648044968085e-05, |
| "loss": 0.2144, |
| "mean_token_accuracy": 0.9393754541873932, |
| "num_tokens": 20076316.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.4856687898089172, |
| "grad_norm": 0.8510144352912903, |
| "learning_rate": 5.2211921021303615e-05, |
| "loss": 0.2236, |
| "mean_token_accuracy": 0.9330400645732879, |
| "num_tokens": 20411336.0, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.49363057324840764, |
| "grad_norm": 0.6923433542251587, |
| "learning_rate": 5.203463120638009e-05, |
| "loss": 0.2388, |
| "mean_token_accuracy": 0.935349416732788, |
| "num_tokens": 20742529.0, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.5015923566878981, |
| "grad_norm": 0.605758547782898, |
| "learning_rate": 5.18546403466745e-05, |
| "loss": 0.1841, |
| "mean_token_accuracy": 0.9448455095291137, |
| "num_tokens": 21075896.0, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.5095541401273885, |
| "grad_norm": 0.7679846882820129, |
| "learning_rate": 5.167197823097856e-05, |
| "loss": 0.1955, |
| "mean_token_accuracy": 0.9422336518764496, |
| "num_tokens": 21410185.0, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.517515923566879, |
| "grad_norm": 0.806277334690094, |
| "learning_rate": 5.148667509018135e-05, |
| "loss": 0.2577, |
| "mean_token_accuracy": 0.9309152662754059, |
| "num_tokens": 21746228.0, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.5254777070063694, |
| "grad_norm": 0.6878032088279724, |
| "learning_rate": 5.129876159226608e-05, |
| "loss": 0.2023, |
| "mean_token_accuracy": 0.9392339050769806, |
| "num_tokens": 22080851.0, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.5334394904458599, |
| "grad_norm": 1.549191951751709, |
| "learning_rate": 5.110826883723449e-05, |
| "loss": 0.2771, |
| "mean_token_accuracy": 0.9251962900161743, |
| "num_tokens": 22416830.0, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.5414012738853503, |
| "grad_norm": 0.526396632194519, |
| "learning_rate": 5.0915228351959706e-05, |
| "loss": 0.2249, |
| "mean_token_accuracy": 0.9363141894340515, |
| "num_tokens": 22749681.0, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.5493630573248408, |
| "grad_norm": 0.7772343158721924, |
| "learning_rate": 5.071967208496856e-05, |
| "loss": 0.2455, |
| "mean_token_accuracy": 0.9330043792724609, |
| "num_tokens": 23084781.0, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.5573248407643312, |
| "grad_norm": 0.6522372961044312, |
| "learning_rate": 5.052163240115397e-05, |
| "loss": 0.2167, |
| "mean_token_accuracy": 0.9410141944885254, |
| "num_tokens": 23421626.0, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.5652866242038217, |
| "grad_norm": 0.8648485541343689, |
| "learning_rate": 5.0321142076418564e-05, |
| "loss": 0.2555, |
| "mean_token_accuracy": 0.9286931216716766, |
| "num_tokens": 23754270.0, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.5732484076433121, |
| "grad_norm": 1.647452712059021, |
| "learning_rate": 5.011823429225019e-05, |
| "loss": 0.2219, |
| "mean_token_accuracy": 0.9380306601524353, |
| "num_tokens": 24090194.0, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.5812101910828026, |
| "grad_norm": 0.7107522487640381, |
| "learning_rate": 4.991294263023029e-05, |
| "loss": 0.2279, |
| "mean_token_accuracy": 0.9340076863765716, |
| "num_tokens": 24424214.0, |
| "step": 365 |
| }, |
| { |
| "epoch": 0.589171974522293, |
| "grad_norm": 1.233316421508789, |
| "learning_rate": 4.970530106647618e-05, |
| "loss": 0.2253, |
| "mean_token_accuracy": 0.9372081577777862, |
| "num_tokens": 24756106.0, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.5971337579617835, |
| "grad_norm": 1.1753926277160645, |
| "learning_rate": 4.949534396601783e-05, |
| "loss": 0.227, |
| "mean_token_accuracy": 0.9371241867542267, |
| "num_tokens": 25089958.0, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.6050955414012739, |
| "grad_norm": 0.528756320476532, |
| "learning_rate": 4.928310607711046e-05, |
| "loss": 0.2395, |
| "mean_token_accuracy": 0.9369891524314881, |
| "num_tokens": 25428370.0, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.6130573248407644, |
| "grad_norm": 0.7281137704849243, |
| "learning_rate": 4.9068622525483675e-05, |
| "loss": 0.2109, |
| "mean_token_accuracy": 0.9409759223461152, |
| "num_tokens": 25764345.0, |
| "step": 385 |
| }, |
| { |
| "epoch": 0.6210191082802548, |
| "grad_norm": 0.5414689779281616, |
| "learning_rate": 4.885192880852802e-05, |
| "loss": 0.1895, |
| "mean_token_accuracy": 0.9459003210067749, |
| "num_tokens": 26103754.0, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.6289808917197452, |
| "grad_norm": 0.9202976226806641, |
| "learning_rate": 4.863306078942015e-05, |
| "loss": 0.2487, |
| "mean_token_accuracy": 0.9336827039718628, |
| "num_tokens": 26437622.0, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.6369426751592356, |
| "grad_norm": 1.2768785953521729, |
| "learning_rate": 4.84120546911874e-05, |
| "loss": 0.2408, |
| "mean_token_accuracy": 0.9370798528194427, |
| "num_tokens": 26770420.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6449044585987261, |
| "grad_norm": 0.9512752890586853, |
| "learning_rate": 4.818894709071284e-05, |
| "loss": 0.224, |
| "mean_token_accuracy": 0.9353628039360047, |
| "num_tokens": 27099112.0, |
| "step": 405 |
| }, |
| { |
| "epoch": 0.6528662420382165, |
| "grad_norm": 0.7952690124511719, |
| "learning_rate": 4.796377491268169e-05, |
| "loss": 0.2106, |
| "mean_token_accuracy": 0.9397002756595612, |
| "num_tokens": 27432075.0, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.660828025477707, |
| "grad_norm": 1.0294992923736572, |
| "learning_rate": 4.773657542347026e-05, |
| "loss": 0.2467, |
| "mean_token_accuracy": 0.9317767977714538, |
| "num_tokens": 27763969.0, |
| "step": 415 |
| }, |
| { |
| "epoch": 0.6687898089171974, |
| "grad_norm": 0.4394127130508423, |
| "learning_rate": 4.7507386224978267e-05, |
| "loss": 0.2117, |
| "mean_token_accuracy": 0.9385966837406159, |
| "num_tokens": 28102253.0, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.6767515923566879, |
| "grad_norm": 0.6836057305335999, |
| "learning_rate": 4.727624524840571e-05, |
| "loss": 0.2174, |
| "mean_token_accuracy": 0.9402537167072296, |
| "num_tokens": 28440233.0, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.6847133757961783, |
| "grad_norm": 0.9111888408660889, |
| "learning_rate": 4.7043190747975144e-05, |
| "loss": 0.188, |
| "mean_token_accuracy": 0.9485961735248566, |
| "num_tokens": 28774168.0, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.6926751592356688, |
| "grad_norm": 0.6586045622825623, |
| "learning_rate": 4.680826129460056e-05, |
| "loss": 0.1972, |
| "mean_token_accuracy": 0.9394764244556427, |
| "num_tokens": 29113264.0, |
| "step": 435 |
| }, |
| { |
| "epoch": 0.7006369426751592, |
| "grad_norm": 1.3775714635849, |
| "learning_rate": 4.6571495769503845e-05, |
| "loss": 0.2096, |
| "mean_token_accuracy": 0.9434356689453125, |
| "num_tokens": 29443242.0, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.7085987261146497, |
| "grad_norm": 0.6877272129058838, |
| "learning_rate": 4.6332933357779854e-05, |
| "loss": 0.2043, |
| "mean_token_accuracy": 0.940096914768219, |
| "num_tokens": 29777629.0, |
| "step": 445 |
| }, |
| { |
| "epoch": 0.7165605095541401, |
| "grad_norm": 0.6645593643188477, |
| "learning_rate": 4.60926135419112e-05, |
| "loss": 0.2071, |
| "mean_token_accuracy": 0.9427264094352722, |
| "num_tokens": 30114597.0, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.7245222929936306, |
| "grad_norm": 0.7703014612197876, |
| "learning_rate": 4.585057609523387e-05, |
| "loss": 0.208, |
| "mean_token_accuracy": 0.9396361827850341, |
| "num_tokens": 30445330.0, |
| "step": 455 |
| }, |
| { |
| "epoch": 0.732484076433121, |
| "grad_norm": 0.6707454323768616, |
| "learning_rate": 4.5606861075354665e-05, |
| "loss": 0.2178, |
| "mean_token_accuracy": 0.9424676895141602, |
| "num_tokens": 30784295.0, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.7404458598726115, |
| "grad_norm": 0.7346827983856201, |
| "learning_rate": 4.536150881752157e-05, |
| "loss": 0.2412, |
| "mean_token_accuracy": 0.9361556649208069, |
| "num_tokens": 31122799.0, |
| "step": 465 |
| }, |
| { |
| "epoch": 0.7484076433121019, |
| "grad_norm": 0.5051635503768921, |
| "learning_rate": 4.511455992794824e-05, |
| "loss": 0.2112, |
| "mean_token_accuracy": 0.9407703936100006, |
| "num_tokens": 31455417.0, |
| "step": 470 |
| } |
| ], |
| "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": 6.59700774333481e+17, |
| "train_batch_size": 24, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
