| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 10.0, |
| "eval_steps": 500, |
| "global_step": 1890, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.7671512730419636, |
| "epoch": 0.05291005291005291, |
| "grad_norm": 14.8125, |
| "learning_rate": 3.2142857142857147e-06, |
| "loss": 1.706784439086914, |
| "mean_token_accuracy": 0.6392517775297165, |
| "num_tokens": 710093.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.4565640933811665, |
| "epoch": 0.10582010582010581, |
| "grad_norm": 6.1875, |
| "learning_rate": 6.785714285714287e-06, |
| "loss": 1.4239604949951172, |
| "mean_token_accuracy": 0.6772899225354194, |
| "num_tokens": 1429271.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.0071680538356305, |
| "epoch": 0.15873015873015872, |
| "grad_norm": 3.109375, |
| "learning_rate": 1.0357142857142859e-05, |
| "loss": 1.048217487335205, |
| "mean_token_accuracy": 0.7393087036907673, |
| "num_tokens": 2149735.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.8510661780834198, |
| "epoch": 0.21164021164021163, |
| "grad_norm": 1.5703125, |
| "learning_rate": 1.3928571428571429e-05, |
| "loss": 0.8619060516357422, |
| "mean_token_accuracy": 0.7765810377895832, |
| "num_tokens": 2868867.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.7774681895971298, |
| "epoch": 0.26455026455026454, |
| "grad_norm": 1.578125, |
| "learning_rate": 1.7500000000000002e-05, |
| "loss": 0.7852191925048828, |
| "mean_token_accuracy": 0.7921242833137512, |
| "num_tokens": 3582476.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.7451388858258724, |
| "epoch": 0.31746031746031744, |
| "grad_norm": 1.5859375, |
| "learning_rate": 1.999986795783372e-05, |
| "loss": 0.748009204864502, |
| "mean_token_accuracy": 0.7999738156795502, |
| "num_tokens": 4287689.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.729080930352211, |
| "epoch": 0.37037037037037035, |
| "grad_norm": 1.4921875, |
| "learning_rate": 1.9997520638548385e-05, |
| "loss": 0.7400224685668946, |
| "mean_token_accuracy": 0.801699859648943, |
| "num_tokens": 5011191.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.7048044994473457, |
| "epoch": 0.42328042328042326, |
| "grad_norm": 1.4453125, |
| "learning_rate": 1.9992239841688418e-05, |
| "loss": 0.7069188117980957, |
| "mean_token_accuracy": 0.8093748845160007, |
| "num_tokens": 5714842.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.6840555652976036, |
| "epoch": 0.47619047619047616, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.9984027116747914e-05, |
| "loss": 0.6863717555999755, |
| "mean_token_accuracy": 0.8138940587639809, |
| "num_tokens": 6424614.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.6870591521263123, |
| "epoch": 0.5291005291005291, |
| "grad_norm": 1.4140625, |
| "learning_rate": 1.997288487350878e-05, |
| "loss": 0.6901006698608398, |
| "mean_token_accuracy": 0.8129441156983376, |
| "num_tokens": 7139778.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.6756624668836594, |
| "epoch": 0.582010582010582, |
| "grad_norm": 1.40625, |
| "learning_rate": 1.9958816381333684e-05, |
| "loss": 0.6771352767944336, |
| "mean_token_accuracy": 0.81553890183568, |
| "num_tokens": 7852540.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.6655688174068928, |
| "epoch": 0.6349206349206349, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.994182576820673e-05, |
| "loss": 0.667414665222168, |
| "mean_token_accuracy": 0.8177205875515938, |
| "num_tokens": 8568056.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.6643485642969609, |
| "epoch": 0.6878306878306878, |
| "grad_norm": 1.296875, |
| "learning_rate": 1.9921918019522238e-05, |
| "loss": 0.6721675872802735, |
| "mean_token_accuracy": 0.816884483397007, |
| "num_tokens": 9276742.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.6590477801859379, |
| "epoch": 0.7407407407407407, |
| "grad_norm": 1.3515625, |
| "learning_rate": 1.9899098976621922e-05, |
| "loss": 0.6590158462524414, |
| "mean_token_accuracy": 0.8199199490249157, |
| "num_tokens": 9990163.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.649939077347517, |
| "epoch": 0.7936507936507936, |
| "grad_norm": 1.4609375, |
| "learning_rate": 1.9873375335080915e-05, |
| "loss": 0.6501528263092041, |
| "mean_token_accuracy": 0.82164012119174, |
| "num_tokens": 10710006.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.6468971997499466, |
| "epoch": 0.8465608465608465, |
| "grad_norm": 1.296875, |
| "learning_rate": 1.9844754642743164e-05, |
| "loss": 0.6529253959655762, |
| "mean_token_accuracy": 0.820851170271635, |
| "num_tokens": 11416822.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.6556476138532161, |
| "epoch": 0.8994708994708994, |
| "grad_norm": 1.296875, |
| "learning_rate": 1.981324529750672e-05, |
| "loss": 0.6604087829589844, |
| "mean_token_accuracy": 0.8195641197264194, |
| "num_tokens": 12123118.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.6358714967966079, |
| "epoch": 0.9523809523809523, |
| "grad_norm": 1.3515625, |
| "learning_rate": 1.9778856544859647e-05, |
| "loss": 0.6365831375122071, |
| "mean_token_accuracy": 0.8241398617625236, |
| "num_tokens": 12831058.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.6277156263589859, |
| "epoch": 1.0052910052910053, |
| "grad_norm": 1.25, |
| "learning_rate": 1.9741598475167173e-05, |
| "loss": 0.6229054927825928, |
| "mean_token_accuracy": 0.827630003541708, |
| "num_tokens": 13547943.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.5929765947163105, |
| "epoch": 1.0582010582010581, |
| "grad_norm": 1.421875, |
| "learning_rate": 1.9701482020711004e-05, |
| "loss": 0.5976824760437012, |
| "mean_token_accuracy": 0.8327139303088188, |
| "num_tokens": 14264772.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.5882044766098261, |
| "epoch": 1.1111111111111112, |
| "grad_norm": 1.1953125, |
| "learning_rate": 1.9658518952481543e-05, |
| "loss": 0.5882174968719482, |
| "mean_token_accuracy": 0.8350941069424153, |
| "num_tokens": 14985565.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.5916924901306629, |
| "epoch": 1.164021164021164, |
| "grad_norm": 1.4140625, |
| "learning_rate": 1.9612721876724054e-05, |
| "loss": 0.5934844017028809, |
| "mean_token_accuracy": 0.8336228899657726, |
| "num_tokens": 15701796.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.5877517074346542, |
| "epoch": 1.216931216931217, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.9564104231239726e-05, |
| "loss": 0.589122200012207, |
| "mean_token_accuracy": 0.8334487572312355, |
| "num_tokens": 16412522.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.5912774313241244, |
| "epoch": 1.2698412698412698, |
| "grad_norm": 1.3359375, |
| "learning_rate": 1.9512680281442745e-05, |
| "loss": 0.5940174102783203, |
| "mean_token_accuracy": 0.83340530321002, |
| "num_tokens": 17126523.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.5930401895195245, |
| "epoch": 1.3227513227513228, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.9458465116174534e-05, |
| "loss": 0.5915824890136718, |
| "mean_token_accuracy": 0.8344384111464024, |
| "num_tokens": 17834934.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.5963230866938829, |
| "epoch": 1.3756613756613756, |
| "grad_norm": 1.296875, |
| "learning_rate": 1.940147464327637e-05, |
| "loss": 0.5983511924743652, |
| "mean_token_accuracy": 0.832802091538906, |
| "num_tokens": 18553413.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.5813015405088663, |
| "epoch": 1.4285714285714286, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.9341725584921712e-05, |
| "loss": 0.5832551956176758, |
| "mean_token_accuracy": 0.8360917493700981, |
| "num_tokens": 19270191.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.5801753208041192, |
| "epoch": 1.4814814814814814, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.9279235472709554e-05, |
| "loss": 0.5805642127990722, |
| "mean_token_accuracy": 0.8362542942166329, |
| "num_tokens": 19976552.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.5879705954343081, |
| "epoch": 1.5343915343915344, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.92140226425203e-05, |
| "loss": 0.5905383110046387, |
| "mean_token_accuracy": 0.8341313540935517, |
| "num_tokens": 20676613.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.5889205444604159, |
| "epoch": 1.5873015873015874, |
| "grad_norm": 1.2578125, |
| "learning_rate": 1.914610622913566e-05, |
| "loss": 0.5913270950317383, |
| "mean_token_accuracy": 0.8349448375403881, |
| "num_tokens": 21385978.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.5768780928105116, |
| "epoch": 1.6402116402116402, |
| "grad_norm": 1.234375, |
| "learning_rate": 1.9075506160624068e-05, |
| "loss": 0.5772612571716309, |
| "mean_token_accuracy": 0.8369500823318958, |
| "num_tokens": 22094336.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.5735550723969937, |
| "epoch": 1.693121693121693, |
| "grad_norm": 1.3203125, |
| "learning_rate": 1.900224315249343e-05, |
| "loss": 0.5721980094909668, |
| "mean_token_accuracy": 0.8380511656403542, |
| "num_tokens": 22798906.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.57716367803514, |
| "epoch": 1.746031746031746, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.8926338701612738e-05, |
| "loss": 0.5769223690032959, |
| "mean_token_accuracy": 0.8371222764253616, |
| "num_tokens": 23511239.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.582606166601181, |
| "epoch": 1.798941798941799, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.884781507990445e-05, |
| "loss": 0.5855788230895996, |
| "mean_token_accuracy": 0.8356470488011837, |
| "num_tokens": 24231185.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.5729806918650866, |
| "epoch": 1.8518518518518519, |
| "grad_norm": 1.21875, |
| "learning_rate": 1.8766695327809463e-05, |
| "loss": 0.5730391502380371, |
| "mean_token_accuracy": 0.8385063141584397, |
| "num_tokens": 24944285.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.5666072227060794, |
| "epoch": 1.9047619047619047, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.8683003247526567e-05, |
| "loss": 0.5674479007720947, |
| "mean_token_accuracy": 0.8392404697835445, |
| "num_tokens": 25670441.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.5809141755104065, |
| "epoch": 1.9576719576719577, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.8596763396028384e-05, |
| "loss": 0.5828475952148438, |
| "mean_token_accuracy": 0.8360860221087932, |
| "num_tokens": 26379786.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.5695893839001656, |
| "epoch": 2.0105820105820107, |
| "grad_norm": 1.359375, |
| "learning_rate": 1.8508001077855876e-05, |
| "loss": 0.5633818626403808, |
| "mean_token_accuracy": 0.8408532224595546, |
| "num_tokens": 27097166.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.5167598329484463, |
| "epoch": 2.0634920634920633, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.8416742337693436e-05, |
| "loss": 0.5171660423278809, |
| "mean_token_accuracy": 0.8513922415673733, |
| "num_tokens": 27818991.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.5240574903786183, |
| "epoch": 2.1164021164021163, |
| "grad_norm": 1.5546875, |
| "learning_rate": 1.8323013952726877e-05, |
| "loss": 0.5252029418945312, |
| "mean_token_accuracy": 0.8488071605563163, |
| "num_tokens": 28524822.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.5178623419255018, |
| "epoch": 2.1693121693121693, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.8226843424786434e-05, |
| "loss": 0.5203696250915527, |
| "mean_token_accuracy": 0.8495552279055119, |
| "num_tokens": 29245634.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.5194813564419747, |
| "epoch": 2.2222222222222223, |
| "grad_norm": 1.25, |
| "learning_rate": 1.8128258972277176e-05, |
| "loss": 0.5169014453887939, |
| "mean_token_accuracy": 0.8509399227797985, |
| "num_tokens": 29967135.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.522173210978508, |
| "epoch": 2.2751322751322753, |
| "grad_norm": 1.25, |
| "learning_rate": 1.8027289521899143e-05, |
| "loss": 0.5245429992675781, |
| "mean_token_accuracy": 0.848798380792141, |
| "num_tokens": 30680274.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.5265159368515014, |
| "epoch": 2.328042328042328, |
| "grad_norm": 1.296875, |
| "learning_rate": 1.7923964700159658e-05, |
| "loss": 0.5277297019958496, |
| "mean_token_accuracy": 0.8482260897755622, |
| "num_tokens": 31386859.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.5307009011507035, |
| "epoch": 2.380952380952381, |
| "grad_norm": 1.359375, |
| "learning_rate": 1.78183148246803e-05, |
| "loss": 0.5284142971038819, |
| "mean_token_accuracy": 0.8480213887989521, |
| "num_tokens": 32111007.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.5215292543172836, |
| "epoch": 2.433862433862434, |
| "grad_norm": 1.203125, |
| "learning_rate": 1.77103708953011e-05, |
| "loss": 0.5223949909210205, |
| "mean_token_accuracy": 0.8494340799748897, |
| "num_tokens": 32819180.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.5321523480117321, |
| "epoch": 2.4867724867724865, |
| "grad_norm": 1.2578125, |
| "learning_rate": 1.7600164584984548e-05, |
| "loss": 0.5305499076843262, |
| "mean_token_accuracy": 0.8471677340567112, |
| "num_tokens": 33523113.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.5240828461945057, |
| "epoch": 2.5396825396825395, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.7487728230522085e-05, |
| "loss": 0.5236708164215088, |
| "mean_token_accuracy": 0.8491978161036968, |
| "num_tokens": 34240357.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.5249651741236449, |
| "epoch": 2.5925925925925926, |
| "grad_norm": 1.21875, |
| "learning_rate": 1.737309482304585e-05, |
| "loss": 0.5244390487670898, |
| "mean_token_accuracy": 0.8490321710705757, |
| "num_tokens": 34958032.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.5262073375284672, |
| "epoch": 2.6455026455026456, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.7256297998348388e-05, |
| "loss": 0.5262768268585205, |
| "mean_token_accuracy": 0.8484946198761463, |
| "num_tokens": 35658658.0, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.5266173977404833, |
| "epoch": 2.6984126984126986, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.7137372027013212e-05, |
| "loss": 0.5282751083374023, |
| "mean_token_accuracy": 0.8476969927549363, |
| "num_tokens": 36364754.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.5260804291814566, |
| "epoch": 2.751322751322751, |
| "grad_norm": 1.25, |
| "learning_rate": 1.7016351804359134e-05, |
| "loss": 0.5255792140960693, |
| "mean_token_accuracy": 0.8494147725403309, |
| "num_tokens": 37074678.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.5435763075947762, |
| "epoch": 2.804232804232804, |
| "grad_norm": 1.4296875, |
| "learning_rate": 1.6893272840201243e-05, |
| "loss": 0.5404428482055664, |
| "mean_token_accuracy": 0.8453655727207661, |
| "num_tokens": 37787128.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.5258786808699369, |
| "epoch": 2.857142857142857, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.6768171248431603e-05, |
| "loss": 0.5282983303070068, |
| "mean_token_accuracy": 0.8478383302688599, |
| "num_tokens": 38504013.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.5149568922817707, |
| "epoch": 2.91005291005291, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.66410837364227e-05, |
| "loss": 0.5157745361328125, |
| "mean_token_accuracy": 0.8509050741791725, |
| "num_tokens": 39212782.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.5198382031172514, |
| "epoch": 2.962962962962963, |
| "grad_norm": 1.296875, |
| "learning_rate": 1.651204759425674e-05, |
| "loss": 0.5191741943359375, |
| "mean_token_accuracy": 0.8499824590981007, |
| "num_tokens": 39931947.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.5168990429490805, |
| "epoch": 3.015873015873016, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.6381100683783984e-05, |
| "loss": 0.5093559265136719, |
| "mean_token_accuracy": 0.8528587318956852, |
| "num_tokens": 40643225.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.4868734121322632, |
| "epoch": 3.068783068783069, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.6248281427513293e-05, |
| "loss": 0.49291515350341797, |
| "mean_token_accuracy": 0.8561817370355129, |
| "num_tokens": 41359577.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.490956873819232, |
| "epoch": 3.121693121693122, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.6113628797338182e-05, |
| "loss": 0.48886570930480955, |
| "mean_token_accuracy": 0.85729940533638, |
| "num_tokens": 42068599.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.47903846949338913, |
| "epoch": 3.1746031746031744, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.597718230310167e-05, |
| "loss": 0.4768198013305664, |
| "mean_token_accuracy": 0.8604256264865399, |
| "num_tokens": 42785175.0, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.4701235756278038, |
| "epoch": 3.2275132275132274, |
| "grad_norm": 1.21875, |
| "learning_rate": 1.5838981981003273e-05, |
| "loss": 0.4688236713409424, |
| "mean_token_accuracy": 0.8626316115260124, |
| "num_tokens": 43504295.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.4784398306161165, |
| "epoch": 3.2804232804232805, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.5699068381851576e-05, |
| "loss": 0.47977123260498045, |
| "mean_token_accuracy": 0.8589689329266548, |
| "num_tokens": 44210262.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.48374257422983646, |
| "epoch": 3.3333333333333335, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.5557482559165777e-05, |
| "loss": 0.4810905456542969, |
| "mean_token_accuracy": 0.8588056646287441, |
| "num_tokens": 44921899.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.47944297045469286, |
| "epoch": 3.386243386243386, |
| "grad_norm": 1.234375, |
| "learning_rate": 1.541426605712975e-05, |
| "loss": 0.480026912689209, |
| "mean_token_accuracy": 0.8588721737265587, |
| "num_tokens": 45631735.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.4891296233981848, |
| "epoch": 3.439153439153439, |
| "grad_norm": 1.515625, |
| "learning_rate": 1.526946089840214e-05, |
| "loss": 0.4879168510437012, |
| "mean_token_accuracy": 0.8568189851939678, |
| "num_tokens": 46339604.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.4852400213479996, |
| "epoch": 3.492063492063492, |
| "grad_norm": 1.2578125, |
| "learning_rate": 1.5123109571786034e-05, |
| "loss": 0.48280906677246094, |
| "mean_token_accuracy": 0.8593833275139332, |
| "num_tokens": 47060193.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.48768885880708696, |
| "epoch": 3.544973544973545, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.4975255019761896e-05, |
| "loss": 0.4901734352111816, |
| "mean_token_accuracy": 0.8567128427326679, |
| "num_tokens": 47768090.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.4818777460604906, |
| "epoch": 3.597883597883598, |
| "grad_norm": 1.25, |
| "learning_rate": 1.4825940625887343e-05, |
| "loss": 0.4794710636138916, |
| "mean_token_accuracy": 0.8597230516374111, |
| "num_tokens": 48486016.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.4782646283507347, |
| "epoch": 3.6507936507936507, |
| "grad_norm": 1.21875, |
| "learning_rate": 1.467521020206753e-05, |
| "loss": 0.4785307884216309, |
| "mean_token_accuracy": 0.8597849570214748, |
| "num_tokens": 49196144.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.48901230953633784, |
| "epoch": 3.7037037037037037, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.4523107975699833e-05, |
| "loss": 0.4887538909912109, |
| "mean_token_accuracy": 0.8576067104935646, |
| "num_tokens": 49912021.0, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.48083703890442847, |
| "epoch": 3.7566137566137567, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.4369678576696633e-05, |
| "loss": 0.47971267700195314, |
| "mean_token_accuracy": 0.8592194318771362, |
| "num_tokens": 50634262.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.47896665148437023, |
| "epoch": 3.8095238095238093, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.4214967024389978e-05, |
| "loss": 0.4767903327941895, |
| "mean_token_accuracy": 0.8601033940911293, |
| "num_tokens": 51343452.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.4764186542481184, |
| "epoch": 3.8624338624338623, |
| "grad_norm": 1.2578125, |
| "learning_rate": 1.4059018714321986e-05, |
| "loss": 0.47649202346801756, |
| "mean_token_accuracy": 0.8595595695078373, |
| "num_tokens": 52046363.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.476634269207716, |
| "epoch": 3.9153439153439153, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.390187940492486e-05, |
| "loss": 0.4764859199523926, |
| "mean_token_accuracy": 0.8600306704640388, |
| "num_tokens": 52755135.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.47171572409570217, |
| "epoch": 3.9682539682539684, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.3743595204094442e-05, |
| "loss": 0.47093925476074217, |
| "mean_token_accuracy": 0.8609667584300041, |
| "num_tokens": 53473566.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.464452663064003, |
| "epoch": 4.021164021164021, |
| "grad_norm": 1.1875, |
| "learning_rate": 1.3584212555661187e-05, |
| "loss": 0.4580575942993164, |
| "mean_token_accuracy": 0.8645686432719231, |
| "num_tokens": 54190492.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.4455810472369194, |
| "epoch": 4.074074074074074, |
| "grad_norm": 1.25, |
| "learning_rate": 1.3423778225762609e-05, |
| "loss": 0.4465456962585449, |
| "mean_token_accuracy": 0.8673712603747845, |
| "num_tokens": 54897936.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.4556675568223, |
| "epoch": 4.1269841269841265, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.3262339289121129e-05, |
| "loss": 0.4551807403564453, |
| "mean_token_accuracy": 0.8652156829833985, |
| "num_tokens": 55609388.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.45422575771808626, |
| "epoch": 4.1798941798941796, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.3099943115231386e-05, |
| "loss": 0.45353970527648924, |
| "mean_token_accuracy": 0.8658270433545112, |
| "num_tokens": 56329299.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.45179531499743464, |
| "epoch": 4.232804232804233, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.2936637354461064e-05, |
| "loss": 0.44892282485961915, |
| "mean_token_accuracy": 0.8666822284460067, |
| "num_tokens": 57042704.0, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.45655727535486224, |
| "epoch": 4.285714285714286, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.2772469924069285e-05, |
| "loss": 0.4565535545349121, |
| "mean_token_accuracy": 0.8650884844362736, |
| "num_tokens": 57755769.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.4504398550838232, |
| "epoch": 4.338624338624339, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.2607488994146703e-05, |
| "loss": 0.44783792495727537, |
| "mean_token_accuracy": 0.8668409429490567, |
| "num_tokens": 58467967.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.45743274949491025, |
| "epoch": 4.391534391534392, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.2441742973481432e-05, |
| "loss": 0.45568046569824217, |
| "mean_token_accuracy": 0.8652968890964985, |
| "num_tokens": 59188579.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.44639028087258337, |
| "epoch": 4.444444444444445, |
| "grad_norm": 1.3515625, |
| "learning_rate": 1.2275280495354889e-05, |
| "loss": 0.44577698707580565, |
| "mean_token_accuracy": 0.8682797506451607, |
| "num_tokens": 59899589.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.4485243815928698, |
| "epoch": 4.497354497354498, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.21081504032718e-05, |
| "loss": 0.4481363773345947, |
| "mean_token_accuracy": 0.8678335674107075, |
| "num_tokens": 60620913.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.4525593984872103, |
| "epoch": 4.550264550264551, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.1940401736628524e-05, |
| "loss": 0.4498861312866211, |
| "mean_token_accuracy": 0.8662019960582257, |
| "num_tokens": 61330675.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.4605423126369715, |
| "epoch": 4.603174603174603, |
| "grad_norm": 1.4140625, |
| "learning_rate": 1.1772083716323878e-05, |
| "loss": 0.4589731216430664, |
| "mean_token_accuracy": 0.8638311237096786, |
| "num_tokens": 62041645.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.44991469383239746, |
| "epoch": 4.656084656084656, |
| "grad_norm": 1.25, |
| "learning_rate": 1.1603245730316724e-05, |
| "loss": 0.447885799407959, |
| "mean_token_accuracy": 0.8659505993127823, |
| "num_tokens": 62750446.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.4469064295291901, |
| "epoch": 4.708994708994709, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.1433937319134512e-05, |
| "loss": 0.44426918029785156, |
| "mean_token_accuracy": 0.868197600543499, |
| "num_tokens": 63460008.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.4514023557305336, |
| "epoch": 4.761904761904762, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.1264208161337091e-05, |
| "loss": 0.45235776901245117, |
| "mean_token_accuracy": 0.8655660562217236, |
| "num_tokens": 64164803.0, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.4533011969178915, |
| "epoch": 4.814814814814815, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.1094108058940005e-05, |
| "loss": 0.4506092548370361, |
| "mean_token_accuracy": 0.865882170945406, |
| "num_tokens": 64878163.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.44160800203680994, |
| "epoch": 4.867724867724868, |
| "grad_norm": 1.296875, |
| "learning_rate": 1.0923686922801537e-05, |
| "loss": 0.4400449752807617, |
| "mean_token_accuracy": 0.868823654204607, |
| "num_tokens": 65599878.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.4407290514558554, |
| "epoch": 4.920634920634921, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.0752994757977862e-05, |
| "loss": 0.4377620697021484, |
| "mean_token_accuracy": 0.8691893480718136, |
| "num_tokens": 66312478.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.45381851457059386, |
| "epoch": 4.973544973544973, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.058208164905052e-05, |
| "loss": 0.45404396057128904, |
| "mean_token_accuracy": 0.8656110934913158, |
| "num_tokens": 67026652.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.4451509706676006, |
| "epoch": 5.026455026455026, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.041099774543059e-05, |
| "loss": 0.43704824447631835, |
| "mean_token_accuracy": 0.8695758722722531, |
| "num_tokens": 67743837.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.4223184697329998, |
| "epoch": 5.079365079365079, |
| "grad_norm": 1.375, |
| "learning_rate": 1.023979324664382e-05, |
| "loss": 0.4231288909912109, |
| "mean_token_accuracy": 0.8729359813034534, |
| "num_tokens": 68459077.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.4201000899076462, |
| "epoch": 5.132275132275132, |
| "grad_norm": 1.3515625, |
| "learning_rate": 1.0068518387601038e-05, |
| "loss": 0.4157362461090088, |
| "mean_token_accuracy": 0.8749549694359302, |
| "num_tokens": 69175886.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.4203161742538214, |
| "epoch": 5.185185185185185, |
| "grad_norm": 1.328125, |
| "learning_rate": 9.897223423858226e-06, |
| "loss": 0.4191565036773682, |
| "mean_token_accuracy": 0.8741719849407673, |
| "num_tokens": 69888377.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.4278976671397686, |
| "epoch": 5.238095238095238, |
| "grad_norm": 1.3125, |
| "learning_rate": 9.725958616870495e-06, |
| "loss": 0.42658100128173826, |
| "mean_token_accuracy": 0.8726167634129525, |
| "num_tokens": 70604584.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.43622247986495494, |
| "epoch": 5.291005291005291, |
| "grad_norm": 1.3359375, |
| "learning_rate": 9.554774219244344e-06, |
| "loss": 0.4340261459350586, |
| "mean_token_accuracy": 0.8702079840004444, |
| "num_tokens": 71321169.0, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.42725917547941206, |
| "epoch": 5.343915343915344, |
| "grad_norm": 1.296875, |
| "learning_rate": 9.38372045999252e-06, |
| "loss": 0.4253455638885498, |
| "mean_token_accuracy": 0.8729547396302223, |
| "num_tokens": 72041175.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.4243607401847839, |
| "epoch": 5.396825396825397, |
| "grad_norm": 1.3828125, |
| "learning_rate": 9.212847529795784e-06, |
| "loss": 0.42325568199157715, |
| "mean_token_accuracy": 0.8731870628893376, |
| "num_tokens": 72749079.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.4303685359656811, |
| "epoch": 5.449735449735449, |
| "grad_norm": 1.2734375, |
| "learning_rate": 9.042205566275952e-06, |
| "loss": 0.4277714729309082, |
| "mean_token_accuracy": 0.8721092127263546, |
| "num_tokens": 73462893.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.43162608668208124, |
| "epoch": 5.502645502645502, |
| "grad_norm": 1.28125, |
| "learning_rate": 8.871844639284479e-06, |
| "loss": 0.4301889419555664, |
| "mean_token_accuracy": 0.8712550736963749, |
| "num_tokens": 74167133.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.43598169907927514, |
| "epoch": 5.555555555555555, |
| "grad_norm": 1.2734375, |
| "learning_rate": 8.701814736210933e-06, |
| "loss": 0.435499382019043, |
| "mean_token_accuracy": 0.8700432546436787, |
| "num_tokens": 74870399.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.4320717755705118, |
| "epoch": 5.608465608465608, |
| "grad_norm": 1.3203125, |
| "learning_rate": 8.532165747315686e-06, |
| "loss": 0.43087425231933596, |
| "mean_token_accuracy": 0.8712112359702587, |
| "num_tokens": 75590782.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.43664140328764917, |
| "epoch": 5.661375661375661, |
| "grad_norm": 1.2734375, |
| "learning_rate": 8.362947451091072e-06, |
| "loss": 0.4324526309967041, |
| "mean_token_accuracy": 0.8710537351667881, |
| "num_tokens": 76305788.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.43034715987741945, |
| "epoch": 5.714285714285714, |
| "grad_norm": 1.2890625, |
| "learning_rate": 8.194209499655364e-06, |
| "loss": 0.4322338581085205, |
| "mean_token_accuracy": 0.8713471658527852, |
| "num_tokens": 77023579.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.4325398612767458, |
| "epoch": 5.767195767195767, |
| "grad_norm": 1.40625, |
| "learning_rate": 8.026001404183831e-06, |
| "loss": 0.4275611400604248, |
| "mean_token_accuracy": 0.8720052778720856, |
| "num_tokens": 77720821.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.43604847937822344, |
| "epoch": 5.8201058201058204, |
| "grad_norm": 1.296875, |
| "learning_rate": 7.85837252038112e-06, |
| "loss": 0.43659210205078125, |
| "mean_token_accuracy": 0.8698216550052166, |
| "num_tokens": 78431671.0, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.4338367357850075, |
| "epoch": 5.8730158730158735, |
| "grad_norm": 1.3125, |
| "learning_rate": 7.69137203399931e-06, |
| "loss": 0.43023014068603516, |
| "mean_token_accuracy": 0.8713279463350773, |
| "num_tokens": 79141393.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.42576855383813383, |
| "epoch": 5.925925925925926, |
| "grad_norm": 1.234375, |
| "learning_rate": 7.525048946405791e-06, |
| "loss": 0.4262782096862793, |
| "mean_token_accuracy": 0.8731129266321659, |
| "num_tokens": 79858356.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.43451742343604566, |
| "epoch": 5.978835978835979, |
| "grad_norm": 1.3046875, |
| "learning_rate": 7.359452060205245e-06, |
| "loss": 0.43216233253479003, |
| "mean_token_accuracy": 0.8712433956563472, |
| "num_tokens": 80566532.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.42183659635484216, |
| "epoch": 6.031746031746032, |
| "grad_norm": 1.28125, |
| "learning_rate": 7.194629964919989e-06, |
| "loss": 0.4154026985168457, |
| "mean_token_accuracy": 0.8752984322607518, |
| "num_tokens": 81283415.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.4100382808595896, |
| "epoch": 6.084656084656085, |
| "grad_norm": 1.2890625, |
| "learning_rate": 7.030631022732781e-06, |
| "loss": 0.40886564254760743, |
| "mean_token_accuracy": 0.8769385173916817, |
| "num_tokens": 81995607.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.4237165842205286, |
| "epoch": 6.137566137566138, |
| "grad_norm": 1.359375, |
| "learning_rate": 6.867503354296392e-06, |
| "loss": 0.42139520645141604, |
| "mean_token_accuracy": 0.8735997013747692, |
| "num_tokens": 82712148.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.42654485553503035, |
| "epoch": 6.190476190476191, |
| "grad_norm": 1.34375, |
| "learning_rate": 6.705294824614004e-06, |
| "loss": 0.4240893363952637, |
| "mean_token_accuracy": 0.8727411277592182, |
| "num_tokens": 83425506.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.4187860660254955, |
| "epoch": 6.243386243386244, |
| "grad_norm": 1.328125, |
| "learning_rate": 6.54405302899463e-06, |
| "loss": 0.41619367599487306, |
| "mean_token_accuracy": 0.8750131532549859, |
| "num_tokens": 84137018.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.4216837890446186, |
| "epoch": 6.296296296296296, |
| "grad_norm": 1.3125, |
| "learning_rate": 6.383825279087699e-06, |
| "loss": 0.41980342864990233, |
| "mean_token_accuracy": 0.8740883760154248, |
| "num_tokens": 84849598.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.4209607470780611, |
| "epoch": 6.349206349206349, |
| "grad_norm": 1.296875, |
| "learning_rate": 6.2246585890008245e-06, |
| "loss": 0.4199337959289551, |
| "mean_token_accuracy": 0.8746874034404755, |
| "num_tokens": 85561962.0, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.41288652531802655, |
| "epoch": 6.402116402116402, |
| "grad_norm": 1.359375, |
| "learning_rate": 6.06659966150491e-06, |
| "loss": 0.4091951370239258, |
| "mean_token_accuracy": 0.8767975211143494, |
| "num_tokens": 86286338.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.41441997736692426, |
| "epoch": 6.455026455026455, |
| "grad_norm": 1.3125, |
| "learning_rate": 5.909694874330601e-06, |
| "loss": 0.4135852813720703, |
| "mean_token_accuracy": 0.8761776715517045, |
| "num_tokens": 87001911.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.41416131891310215, |
| "epoch": 6.507936507936508, |
| "grad_norm": 1.28125, |
| "learning_rate": 5.753990266560105e-06, |
| "loss": 0.4115488052368164, |
| "mean_token_accuracy": 0.8769605971872807, |
| "num_tokens": 87714571.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.4153778996318579, |
| "epoch": 6.560846560846561, |
| "grad_norm": 1.375, |
| "learning_rate": 5.599531525118373e-06, |
| "loss": 0.4159687519073486, |
| "mean_token_accuracy": 0.8752469770610333, |
| "num_tokens": 88417674.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.4190424919128418, |
| "epoch": 6.613756613756614, |
| "grad_norm": 1.3203125, |
| "learning_rate": 5.446363971367638e-06, |
| "loss": 0.4144894599914551, |
| "mean_token_accuracy": 0.875180634111166, |
| "num_tokens": 89123522.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.42192798517644403, |
| "epoch": 6.666666666666667, |
| "grad_norm": 1.34375, |
| "learning_rate": 5.29453254780917e-06, |
| "loss": 0.4181344032287598, |
| "mean_token_accuracy": 0.8751389637589455, |
| "num_tokens": 89838831.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.4147109277546406, |
| "epoch": 6.71957671957672, |
| "grad_norm": 1.2578125, |
| "learning_rate": 5.1440818048962575e-06, |
| "loss": 0.4149214744567871, |
| "mean_token_accuracy": 0.875646834820509, |
| "num_tokens": 90546066.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.41611550822854043, |
| "epoch": 6.772486772486772, |
| "grad_norm": 1.359375, |
| "learning_rate": 4.995055887962167e-06, |
| "loss": 0.41540231704711916, |
| "mean_token_accuracy": 0.875544361025095, |
| "num_tokens": 91257301.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.4205140460282564, |
| "epoch": 6.825396825396825, |
| "grad_norm": 1.375, |
| "learning_rate": 4.8474985242669965e-06, |
| "loss": 0.4182106018066406, |
| "mean_token_accuracy": 0.8747602418065071, |
| "num_tokens": 91961075.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.4291623383760452, |
| "epoch": 6.878306878306878, |
| "grad_norm": 1.3984375, |
| "learning_rate": 4.701453010167217e-06, |
| "loss": 0.42667760848999026, |
| "mean_token_accuracy": 0.8727425791323185, |
| "num_tokens": 92677769.0, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.41312588155269625, |
| "epoch": 6.931216931216931, |
| "grad_norm": 1.3359375, |
| "learning_rate": 4.556962198411631e-06, |
| "loss": 0.4099729537963867, |
| "mean_token_accuracy": 0.8769033893942833, |
| "num_tokens": 93390418.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.4084371831268072, |
| "epoch": 6.984126984126984, |
| "grad_norm": 1.3671875, |
| "learning_rate": 4.414068485567473e-06, |
| "loss": 0.4083920955657959, |
| "mean_token_accuracy": 0.8769270032644272, |
| "num_tokens": 94107638.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.414901152998209, |
| "epoch": 7.037037037037037, |
| "grad_norm": 1.2890625, |
| "learning_rate": 4.272813799580417e-06, |
| "loss": 0.4091632843017578, |
| "mean_token_accuracy": 0.8766109876334667, |
| "num_tokens": 94829458.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.4114344149827957, |
| "epoch": 7.08994708994709, |
| "grad_norm": 1.296875, |
| "learning_rate": 4.1332395874720474e-06, |
| "loss": 0.40756850242614745, |
| "mean_token_accuracy": 0.8767555914819241, |
| "num_tokens": 95540140.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.4090781681239605, |
| "epoch": 7.142857142857143, |
| "grad_norm": 1.265625, |
| "learning_rate": 3.9953868031784415e-06, |
| "loss": 0.4089008331298828, |
| "mean_token_accuracy": 0.876671238988638, |
| "num_tokens": 96254434.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.4147778756916523, |
| "epoch": 7.195767195767195, |
| "grad_norm": 1.328125, |
| "learning_rate": 3.8592958955334636e-06, |
| "loss": 0.4134193420410156, |
| "mean_token_accuracy": 0.8760911405086518, |
| "num_tokens": 96966014.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.40922857597470286, |
| "epoch": 7.248677248677248, |
| "grad_norm": 1.34375, |
| "learning_rate": 3.725006796400231e-06, |
| "loss": 0.40600833892822263, |
| "mean_token_accuracy": 0.8775057464838028, |
| "num_tokens": 97677188.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.41339924447238446, |
| "epoch": 7.301587301587301, |
| "grad_norm": 1.421875, |
| "learning_rate": 3.5925589089542954e-06, |
| "loss": 0.4097587585449219, |
| "mean_token_accuracy": 0.875959849357605, |
| "num_tokens": 98391677.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.41218244582414626, |
| "epoch": 7.354497354497354, |
| "grad_norm": 1.3046875, |
| "learning_rate": 3.461991096121955e-06, |
| "loss": 0.4099538326263428, |
| "mean_token_accuracy": 0.8761332534253597, |
| "num_tokens": 99105789.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.410692673176527, |
| "epoch": 7.407407407407407, |
| "grad_norm": 1.3203125, |
| "learning_rate": 3.3333416691770402e-06, |
| "loss": 0.4073766231536865, |
| "mean_token_accuracy": 0.8764613293111324, |
| "num_tokens": 99819774.0, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.4076389245688915, |
| "epoch": 7.4603174603174605, |
| "grad_norm": 1.3046875, |
| "learning_rate": 3.206648376499637e-06, |
| "loss": 0.4059453964233398, |
| "mean_token_accuracy": 0.8779646269977093, |
| "num_tokens": 100540345.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.416654784604907, |
| "epoch": 7.5132275132275135, |
| "grad_norm": 1.3046875, |
| "learning_rate": 3.0819483924999183e-06, |
| "loss": 0.41609792709350585, |
| "mean_token_accuracy": 0.8747843399643898, |
| "num_tokens": 101261007.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.41376331858336923, |
| "epoch": 7.5661375661375665, |
| "grad_norm": 1.2890625, |
| "learning_rate": 2.959278306710409e-06, |
| "loss": 0.40964488983154296, |
| "mean_token_accuracy": 0.8766733475029469, |
| "num_tokens": 101972692.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.4122611477971077, |
| "epoch": 7.619047619047619, |
| "grad_norm": 1.3125, |
| "learning_rate": 2.838674113049884e-06, |
| "loss": 0.4077149391174316, |
| "mean_token_accuracy": 0.8768246524035931, |
| "num_tokens": 102684836.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.42518827989697455, |
| "epoch": 7.671957671957672, |
| "grad_norm": 1.375, |
| "learning_rate": 2.7201711992619873e-06, |
| "loss": 0.42249202728271484, |
| "mean_token_accuracy": 0.8743405237793922, |
| "num_tokens": 103404477.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.41746283397078515, |
| "epoch": 7.724867724867725, |
| "grad_norm": 1.328125, |
| "learning_rate": 2.6038043365317643e-06, |
| "loss": 0.41622037887573243, |
| "mean_token_accuracy": 0.8755017273128033, |
| "num_tokens": 104111178.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.40634516142308713, |
| "epoch": 7.777777777777778, |
| "grad_norm": 1.3125, |
| "learning_rate": 2.489607669283088e-06, |
| "loss": 0.4032726287841797, |
| "mean_token_accuracy": 0.8784879930317402, |
| "num_tokens": 104820924.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.4181860238313675, |
| "epoch": 7.830687830687831, |
| "grad_norm": 1.3046875, |
| "learning_rate": 2.3776147051599518e-06, |
| "loss": 0.4165492057800293, |
| "mean_token_accuracy": 0.8751473076641559, |
| "num_tokens": 105525661.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.4215067997574806, |
| "epoch": 7.883597883597884, |
| "grad_norm": 1.296875, |
| "learning_rate": 2.2678583051946733e-06, |
| "loss": 0.4215100288391113, |
| "mean_token_accuracy": 0.8738829605281353, |
| "num_tokens": 106244487.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.422141370177269, |
| "epoch": 7.936507936507937, |
| "grad_norm": 1.4140625, |
| "learning_rate": 2.1603706741657803e-06, |
| "loss": 0.41953001022338865, |
| "mean_token_accuracy": 0.8739961586892605, |
| "num_tokens": 106950025.0, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.41841135509312155, |
| "epoch": 7.98941798941799, |
| "grad_norm": 1.3125, |
| "learning_rate": 2.055183351148463e-06, |
| "loss": 0.41535396575927735, |
| "mean_token_accuracy": 0.8752631053328515, |
| "num_tokens": 107659233.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.4093180373311043, |
| "epoch": 8.042328042328043, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.9523272002603745e-06, |
| "loss": 0.4061167240142822, |
| "mean_token_accuracy": 0.8781525194644928, |
| "num_tokens": 108374173.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.3999630343168974, |
| "epoch": 8.095238095238095, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.8518324016054601e-06, |
| "loss": 0.39598617553710935, |
| "mean_token_accuracy": 0.8807546623051167, |
| "num_tokens": 109085583.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.4098013911396265, |
| "epoch": 8.148148148148149, |
| "grad_norm": 1.3203125, |
| "learning_rate": 1.7537284424184852e-06, |
| "loss": 0.40534296035766604, |
| "mean_token_accuracy": 0.8775141790509224, |
| "num_tokens": 109801232.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.41676906384527684, |
| "epoch": 8.201058201058201, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.658044108412905e-06, |
| "loss": 0.41460742950439455, |
| "mean_token_accuracy": 0.8753950335085392, |
| "num_tokens": 110516150.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.4116453532129526, |
| "epoch": 8.253968253968253, |
| "grad_norm": 1.3515625, |
| "learning_rate": 1.564807475334509e-06, |
| "loss": 0.4131956100463867, |
| "mean_token_accuracy": 0.8766112357378006, |
| "num_tokens": 111232078.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.41225610859692097, |
| "epoch": 8.306878306878307, |
| "grad_norm": 1.3359375, |
| "learning_rate": 1.4740459007234498e-06, |
| "loss": 0.41111059188842775, |
| "mean_token_accuracy": 0.8763026840984821, |
| "num_tokens": 111950711.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.41501538529992105, |
| "epoch": 8.359788359788359, |
| "grad_norm": 1.3203125, |
| "learning_rate": 1.3857860158869696e-06, |
| "loss": 0.4131760597229004, |
| "mean_token_accuracy": 0.8759730905294418, |
| "num_tokens": 112658277.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.41387021318078043, |
| "epoch": 8.412698412698413, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.3000537180852213e-06, |
| "loss": 0.4116035461425781, |
| "mean_token_accuracy": 0.8768503859639167, |
| "num_tokens": 113378303.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.4176288653165102, |
| "epoch": 8.465608465608465, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.2168741629324944e-06, |
| "loss": 0.4141057014465332, |
| "mean_token_accuracy": 0.8759131275117398, |
| "num_tokens": 114087597.0, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.40503756925463674, |
| "epoch": 8.518518518518519, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.1362717570160387e-06, |
| "loss": 0.40210609436035155, |
| "mean_token_accuracy": 0.8788188330829143, |
| "num_tokens": 114803563.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.4061105038970709, |
| "epoch": 8.571428571428571, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.0582701507346582e-06, |
| "loss": 0.4021212100982666, |
| "mean_token_accuracy": 0.878644997626543, |
| "num_tokens": 115519221.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.4147233605384827, |
| "epoch": 8.624338624338625, |
| "grad_norm": 1.25, |
| "learning_rate": 9.828922313592227e-07, |
| "loss": 0.410779333114624, |
| "mean_token_accuracy": 0.876928336918354, |
| "num_tokens": 116238487.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.4097800888121128, |
| "epoch": 8.677248677248677, |
| "grad_norm": 1.34375, |
| "learning_rate": 9.10160116317047e-07, |
| "loss": 0.40713653564453123, |
| "mean_token_accuracy": 0.8772961221635341, |
| "num_tokens": 116956616.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.416746861487627, |
| "epoch": 8.73015873015873, |
| "grad_norm": 1.34375, |
| "learning_rate": 8.400951467022067e-07, |
| "loss": 0.41678814888000487, |
| "mean_token_accuracy": 0.8752295158803463, |
| "num_tokens": 117662184.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.4065455045551062, |
| "epoch": 8.783068783068783, |
| "grad_norm": 1.3203125, |
| "learning_rate": 7.727178810136094e-07, |
| "loss": 0.40470399856567385, |
| "mean_token_accuracy": 0.878332145512104, |
| "num_tokens": 118365440.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.4117112413048744, |
| "epoch": 8.835978835978835, |
| "grad_norm": 1.2734375, |
| "learning_rate": 7.080480891227126e-07, |
| "loss": 0.40632357597351076, |
| "mean_token_accuracy": 0.8775922633707524, |
| "num_tokens": 119078464.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.41403592862188815, |
| "epoch": 8.88888888888889, |
| "grad_norm": 1.265625, |
| "learning_rate": 6.461047464726433e-07, |
| "loss": 0.41112241744995115, |
| "mean_token_accuracy": 0.8770215392112732, |
| "num_tokens": 119792226.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.41094372384250166, |
| "epoch": 8.941798941798941, |
| "grad_norm": 1.3359375, |
| "learning_rate": 5.869060285104089e-07, |
| "loss": 0.4097023010253906, |
| "mean_token_accuracy": 0.8770746536552906, |
| "num_tokens": 120500468.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.40498077906668184, |
| "epoch": 8.994708994708995, |
| "grad_norm": 1.28125, |
| "learning_rate": 5.304693053538434e-07, |
| "loss": 0.4013853073120117, |
| "mean_token_accuracy": 0.8787798292934894, |
| "num_tokens": 121205368.0, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.39938552118837833, |
| "epoch": 9.047619047619047, |
| "grad_norm": 1.265625, |
| "learning_rate": 4.768111366948736e-07, |
| "loss": 0.39675068855285645, |
| "mean_token_accuracy": 0.8801792360842228, |
| "num_tokens": 121917493.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.4145715493708849, |
| "epoch": 9.100529100529101, |
| "grad_norm": 1.359375, |
| "learning_rate": 4.2594726694055397e-07, |
| "loss": 0.4126988410949707, |
| "mean_token_accuracy": 0.8764167346060276, |
| "num_tokens": 122643710.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.40943403616547586, |
| "epoch": 9.153439153439153, |
| "grad_norm": 1.265625, |
| "learning_rate": 3.778926205933342e-07, |
| "loss": 0.406523323059082, |
| "mean_token_accuracy": 0.8777760624885559, |
| "num_tokens": 123357455.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.4116696141660213, |
| "epoch": 9.206349206349206, |
| "grad_norm": 1.3125, |
| "learning_rate": 3.3266129787191015e-07, |
| "loss": 0.4094538688659668, |
| "mean_token_accuracy": 0.8771643012762069, |
| "num_tokens": 124071350.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.4071599967777729, |
| "epoch": 9.25925925925926, |
| "grad_norm": 1.3671875, |
| "learning_rate": 2.9026657057392715e-07, |
| "loss": 0.40432915687561033, |
| "mean_token_accuracy": 0.8791363179683686, |
| "num_tokens": 124788814.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.40564643740653994, |
| "epoch": 9.312169312169312, |
| "grad_norm": 1.34375, |
| "learning_rate": 2.507208781817638e-07, |
| "loss": 0.4046478271484375, |
| "mean_token_accuracy": 0.8778819508850575, |
| "num_tokens": 125504162.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.4061614666134119, |
| "epoch": 9.365079365079366, |
| "grad_norm": 1.3203125, |
| "learning_rate": 2.1403582421253398e-07, |
| "loss": 0.40030655860900877, |
| "mean_token_accuracy": 0.8788155451416969, |
| "num_tokens": 126213099.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.40765208303928374, |
| "epoch": 9.417989417989418, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.8022217281336908e-07, |
| "loss": 0.4054518222808838, |
| "mean_token_accuracy": 0.8777305997908116, |
| "num_tokens": 126934123.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.410344659909606, |
| "epoch": 9.470899470899472, |
| "grad_norm": 1.296875, |
| "learning_rate": 1.4928984560301006e-07, |
| "loss": 0.4070878028869629, |
| "mean_token_accuracy": 0.8774436771869659, |
| "num_tokens": 127647228.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.41973356865346434, |
| "epoch": 9.523809523809524, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.212479187605897e-07, |
| "loss": 0.41802120208740234, |
| "mean_token_accuracy": 0.8751970149576664, |
| "num_tokens": 128350815.0, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.41576062627136706, |
| "epoch": 9.576719576719576, |
| "grad_norm": 1.34375, |
| "learning_rate": 9.610462036249823e-08, |
| "loss": 0.41389904022216795, |
| "mean_token_accuracy": 0.8757011353969574, |
| "num_tokens": 129070776.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.40447696633636954, |
| "epoch": 9.62962962962963, |
| "grad_norm": 1.3359375, |
| "learning_rate": 7.386732796810014e-08, |
| "loss": 0.4027875900268555, |
| "mean_token_accuracy": 0.8785097613930702, |
| "num_tokens": 129784905.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.4036568459123373, |
| "epoch": 9.682539682539682, |
| "grad_norm": 1.3515625, |
| "learning_rate": 5.454256645500477e-08, |
| "loss": 0.401032018661499, |
| "mean_token_accuracy": 0.8786628104746341, |
| "num_tokens": 130492958.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.41443242095410826, |
| "epoch": 9.735449735449736, |
| "grad_norm": 1.3671875, |
| "learning_rate": 3.8136006104533276e-08, |
| "loss": 0.4110685348510742, |
| "mean_token_accuracy": 0.8764477990567684, |
| "num_tokens": 131214260.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.4086302760988474, |
| "epoch": 9.788359788359788, |
| "grad_norm": 1.3828125, |
| "learning_rate": 2.4652460937941714e-08, |
| "loss": 0.4072721481323242, |
| "mean_token_accuracy": 0.8769519530236721, |
| "num_tokens": 131927722.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.40928103737533095, |
| "epoch": 9.841269841269842, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.4095887303888778e-08, |
| "loss": 0.40800299644470217, |
| "mean_token_accuracy": 0.8766087085008621, |
| "num_tokens": 132622563.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.4112366572022438, |
| "epoch": 9.894179894179894, |
| "grad_norm": 1.3359375, |
| "learning_rate": 6.469382717563255e-09, |
| "loss": 0.41067938804626464, |
| "mean_token_accuracy": 0.8762929260730743, |
| "num_tokens": 133332199.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.405778194591403, |
| "epoch": 9.947089947089948, |
| "grad_norm": 1.3359375, |
| "learning_rate": 1.775184951806619e-09, |
| "loss": 0.4031084060668945, |
| "mean_token_accuracy": 0.8784385807812214, |
| "num_tokens": 134041278.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.41113952547311783, |
| "epoch": 10.0, |
| "grad_norm": 1.3359375, |
| "learning_rate": 1.4671380509367893e-11, |
| "loss": 0.4097463607788086, |
| "mean_token_accuracy": 0.8767766036093235, |
| "num_tokens": 134755992.0, |
| "step": 1890 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 1890, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.4340057784875745e+18, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
