| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 234, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.2969624996185303, |
| "epoch": 0.012861736334405145, |
| "grad_norm": 1.222815990447998, |
| "learning_rate": 0.0, |
| "loss": 2.025021553039551, |
| "mean_token_accuracy": 0.5397194549441338, |
| "num_tokens": 128512.0, |
| "step": 1 |
| }, |
| { |
| "entropy": 1.3495921194553375, |
| "epoch": 0.02572347266881029, |
| "grad_norm": 1.2900104522705078, |
| "learning_rate": 4.166666666666667e-06, |
| "loss": 2.113243341445923, |
| "mean_token_accuracy": 0.5239669531583786, |
| "num_tokens": 258439.0, |
| "step": 2 |
| }, |
| { |
| "entropy": 1.4065107256174088, |
| "epoch": 0.03858520900321544, |
| "grad_norm": 1.1876918077468872, |
| "learning_rate": 8.333333333333334e-06, |
| "loss": 2.0984230041503906, |
| "mean_token_accuracy": 0.525254875421524, |
| "num_tokens": 387958.0, |
| "step": 3 |
| }, |
| { |
| "entropy": 1.4473730325698853, |
| "epoch": 0.05144694533762058, |
| "grad_norm": 0.995452880859375, |
| "learning_rate": 1.25e-05, |
| "loss": 2.0187673568725586, |
| "mean_token_accuracy": 0.534252941608429, |
| "num_tokens": 516779.0, |
| "step": 4 |
| }, |
| { |
| "entropy": 1.626717284321785, |
| "epoch": 0.06430868167202572, |
| "grad_norm": 0.6222253441810608, |
| "learning_rate": 1.6666666666666667e-05, |
| "loss": 1.9110387563705444, |
| "mean_token_accuracy": 0.5435223057866096, |
| "num_tokens": 645829.0, |
| "step": 5 |
| }, |
| { |
| "entropy": 1.810067042708397, |
| "epoch": 0.07717041800643087, |
| "grad_norm": 0.42118003964424133, |
| "learning_rate": 2.0833333333333336e-05, |
| "loss": 1.8789349794387817, |
| "mean_token_accuracy": 0.5466088876128197, |
| "num_tokens": 774435.0, |
| "step": 6 |
| }, |
| { |
| "entropy": 1.9707088768482208, |
| "epoch": 0.09003215434083602, |
| "grad_norm": 0.4645366072654724, |
| "learning_rate": 2.5e-05, |
| "loss": 1.8074758052825928, |
| "mean_token_accuracy": 0.5567064955830574, |
| "num_tokens": 903545.0, |
| "step": 7 |
| }, |
| { |
| "entropy": 1.9890314191579819, |
| "epoch": 0.10289389067524116, |
| "grad_norm": 0.5461070537567139, |
| "learning_rate": 2.916666666666667e-05, |
| "loss": 1.7440648078918457, |
| "mean_token_accuracy": 0.5663307011127472, |
| "num_tokens": 1033070.0, |
| "step": 8 |
| }, |
| { |
| "entropy": 1.968129649758339, |
| "epoch": 0.1157556270096463, |
| "grad_norm": 0.5405251383781433, |
| "learning_rate": 3.3333333333333335e-05, |
| "loss": 1.7069363594055176, |
| "mean_token_accuracy": 0.5683588162064552, |
| "num_tokens": 1162755.0, |
| "step": 9 |
| }, |
| { |
| "entropy": 1.8345413953065872, |
| "epoch": 0.12861736334405144, |
| "grad_norm": 0.42177239060401917, |
| "learning_rate": 3.7500000000000003e-05, |
| "loss": 1.65086030960083, |
| "mean_token_accuracy": 0.5773059874773026, |
| "num_tokens": 1290164.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.6431169360876083, |
| "epoch": 0.1414790996784566, |
| "grad_norm": 0.31144559383392334, |
| "learning_rate": 4.166666666666667e-05, |
| "loss": 1.582027792930603, |
| "mean_token_accuracy": 0.5872198939323425, |
| "num_tokens": 1419501.0, |
| "step": 11 |
| }, |
| { |
| "entropy": 1.544620230793953, |
| "epoch": 0.15434083601286175, |
| "grad_norm": 0.27825888991355896, |
| "learning_rate": 4.5833333333333334e-05, |
| "loss": 1.5446865558624268, |
| "mean_token_accuracy": 0.594700962305069, |
| "num_tokens": 1547996.0, |
| "step": 12 |
| }, |
| { |
| "entropy": 1.4267793744802475, |
| "epoch": 0.16720257234726688, |
| "grad_norm": 0.2835897207260132, |
| "learning_rate": 5e-05, |
| "loss": 1.509974479675293, |
| "mean_token_accuracy": 0.6006790399551392, |
| "num_tokens": 1676937.0, |
| "step": 13 |
| }, |
| { |
| "entropy": 1.3972271531820297, |
| "epoch": 0.18006430868167203, |
| "grad_norm": 0.25854921340942383, |
| "learning_rate": 4.9997496794168726e-05, |
| "loss": 1.491778016090393, |
| "mean_token_accuracy": 0.6045994758605957, |
| "num_tokens": 1806013.0, |
| "step": 14 |
| }, |
| { |
| "entropy": 1.4395506381988525, |
| "epoch": 0.19292604501607716, |
| "grad_norm": 0.22095996141433716, |
| "learning_rate": 4.998998767795805e-05, |
| "loss": 1.4883487224578857, |
| "mean_token_accuracy": 0.6050755307078362, |
| "num_tokens": 1935387.0, |
| "step": 15 |
| }, |
| { |
| "entropy": 1.4550755470991135, |
| "epoch": 0.2057877813504823, |
| "grad_norm": 0.24221676588058472, |
| "learning_rate": 4.9977474155117045e-05, |
| "loss": 1.4858245849609375, |
| "mean_token_accuracy": 0.6017315089702606, |
| "num_tokens": 2065100.0, |
| "step": 16 |
| }, |
| { |
| "entropy": 1.4459427893161774, |
| "epoch": 0.21864951768488747, |
| "grad_norm": 0.25146999955177307, |
| "learning_rate": 4.995995873155958e-05, |
| "loss": 1.4274994134902954, |
| "mean_token_accuracy": 0.6151479333639145, |
| "num_tokens": 2194685.0, |
| "step": 17 |
| }, |
| { |
| "entropy": 1.4442210048437119, |
| "epoch": 0.2315112540192926, |
| "grad_norm": 0.2271786779165268, |
| "learning_rate": 4.99374449148625e-05, |
| "loss": 1.4092518091201782, |
| "mean_token_accuracy": 0.6148851290345192, |
| "num_tokens": 2324963.0, |
| "step": 18 |
| }, |
| { |
| "entropy": 1.4254360347986221, |
| "epoch": 0.24437299035369775, |
| "grad_norm": 0.18994390964508057, |
| "learning_rate": 4.9909937213563165e-05, |
| "loss": 1.3825007677078247, |
| "mean_token_accuracy": 0.6199355497956276, |
| "num_tokens": 2454343.0, |
| "step": 19 |
| }, |
| { |
| "entropy": 1.3843167871236801, |
| "epoch": 0.2572347266881029, |
| "grad_norm": 0.18420229852199554, |
| "learning_rate": 4.987744113625665e-05, |
| "loss": 1.3471739292144775, |
| "mean_token_accuracy": 0.6283371672034264, |
| "num_tokens": 2584405.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.3585271686315536, |
| "epoch": 0.27009646302250806, |
| "grad_norm": 0.18983817100524902, |
| "learning_rate": 4.9839963190492576e-05, |
| "loss": 1.3283567428588867, |
| "mean_token_accuracy": 0.6319968476891518, |
| "num_tokens": 2709555.0, |
| "step": 21 |
| }, |
| { |
| "entropy": 1.3483584225177765, |
| "epoch": 0.2829581993569132, |
| "grad_norm": 0.18332050740718842, |
| "learning_rate": 4.979751088147192e-05, |
| "loss": 1.33201003074646, |
| "mean_token_accuracy": 0.6298539489507675, |
| "num_tokens": 2839191.0, |
| "step": 22 |
| }, |
| { |
| "entropy": 1.334426462650299, |
| "epoch": 0.2958199356913183, |
| "grad_norm": 0.16615396738052368, |
| "learning_rate": 4.975009271054409e-05, |
| "loss": 1.3058710098266602, |
| "mean_token_accuracy": 0.6357970610260963, |
| "num_tokens": 2968625.0, |
| "step": 23 |
| }, |
| { |
| "entropy": 1.3269707262516022, |
| "epoch": 0.3086816720257235, |
| "grad_norm": 0.16659598052501678, |
| "learning_rate": 4.969771817350445e-05, |
| "loss": 1.298633337020874, |
| "mean_token_accuracy": 0.6390289217233658, |
| "num_tokens": 3098193.0, |
| "step": 24 |
| }, |
| { |
| "entropy": 1.3264441937208176, |
| "epoch": 0.3215434083601286, |
| "grad_norm": 0.18099723756313324, |
| "learning_rate": 4.9640397758692715e-05, |
| "loss": 1.2919727563858032, |
| "mean_token_accuracy": 0.6379936411976814, |
| "num_tokens": 3226304.0, |
| "step": 25 |
| }, |
| { |
| "entropy": 1.2971351444721222, |
| "epoch": 0.33440514469453375, |
| "grad_norm": 0.16978149116039276, |
| "learning_rate": 4.957814294489261e-05, |
| "loss": 1.283421277999878, |
| "mean_token_accuracy": 0.6396878883242607, |
| "num_tokens": 3356150.0, |
| "step": 26 |
| }, |
| { |
| "entropy": 1.2420216798782349, |
| "epoch": 0.34726688102893893, |
| "grad_norm": 0.15654370188713074, |
| "learning_rate": 4.9510966199033174e-05, |
| "loss": 1.2411205768585205, |
| "mean_token_accuracy": 0.6497687473893166, |
| "num_tokens": 3485881.0, |
| "step": 27 |
| }, |
| { |
| "entropy": 1.2905446141958237, |
| "epoch": 0.36012861736334406, |
| "grad_norm": 0.15767726302146912, |
| "learning_rate": 4.943888097369216e-05, |
| "loss": 1.2969235181808472, |
| "mean_token_accuracy": 0.6399867981672287, |
| "num_tokens": 3614360.0, |
| "step": 28 |
| }, |
| { |
| "entropy": 1.2797959297895432, |
| "epoch": 0.3729903536977492, |
| "grad_norm": 0.1609920859336853, |
| "learning_rate": 4.936190170440208e-05, |
| "loss": 1.281036615371704, |
| "mean_token_accuracy": 0.6414945125579834, |
| "num_tokens": 3742757.0, |
| "step": 29 |
| }, |
| { |
| "entropy": 1.2370718121528625, |
| "epoch": 0.3858520900321543, |
| "grad_norm": 0.15289531648159027, |
| "learning_rate": 4.928004380675941e-05, |
| "loss": 1.2397706508636475, |
| "mean_token_accuracy": 0.6485739275813103, |
| "num_tokens": 3872361.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.2661651074886322, |
| "epoch": 0.3987138263665595, |
| "grad_norm": 0.15539875626564026, |
| "learning_rate": 4.9193323673337476e-05, |
| "loss": 1.2545585632324219, |
| "mean_token_accuracy": 0.646148107945919, |
| "num_tokens": 4001029.0, |
| "step": 31 |
| }, |
| { |
| "entropy": 1.259293794631958, |
| "epoch": 0.4115755627009646, |
| "grad_norm": 0.15602311491966248, |
| "learning_rate": 4.910175867040377e-05, |
| "loss": 1.2386581897735596, |
| "mean_token_accuracy": 0.6532392650842667, |
| "num_tokens": 4129746.0, |
| "step": 32 |
| }, |
| { |
| "entropy": 1.2246208488941193, |
| "epoch": 0.42443729903536975, |
| "grad_norm": 0.15040729939937592, |
| "learning_rate": 4.9005367134442235e-05, |
| "loss": 1.2012075185775757, |
| "mean_token_accuracy": 0.6605038940906525, |
| "num_tokens": 4258908.0, |
| "step": 33 |
| }, |
| { |
| "entropy": 1.2688497006893158, |
| "epoch": 0.43729903536977494, |
| "grad_norm": 0.14351816475391388, |
| "learning_rate": 4.890416836848127e-05, |
| "loss": 1.2468310594558716, |
| "mean_token_accuracy": 0.6454005613923073, |
| "num_tokens": 4388837.0, |
| "step": 34 |
| }, |
| { |
| "entropy": 1.2282109707593918, |
| "epoch": 0.45016077170418006, |
| "grad_norm": 0.14232277870178223, |
| "learning_rate": 4.8798182638228166e-05, |
| "loss": 1.2170629501342773, |
| "mean_token_accuracy": 0.6531817615032196, |
| "num_tokens": 4518130.0, |
| "step": 35 |
| }, |
| { |
| "entropy": 1.2145698070526123, |
| "epoch": 0.4630225080385852, |
| "grad_norm": 0.14945274591445923, |
| "learning_rate": 4.868743116801074e-05, |
| "loss": 1.2234522104263306, |
| "mean_token_accuracy": 0.6538028195500374, |
| "num_tokens": 4647727.0, |
| "step": 36 |
| }, |
| { |
| "entropy": 1.2184867858886719, |
| "epoch": 0.4758842443729904, |
| "grad_norm": 0.14716529846191406, |
| "learning_rate": 4.857193613652711e-05, |
| "loss": 1.2205889225006104, |
| "mean_token_accuracy": 0.6552813798189163, |
| "num_tokens": 4776814.0, |
| "step": 37 |
| }, |
| { |
| "entropy": 1.2501244097948074, |
| "epoch": 0.4887459807073955, |
| "grad_norm": 0.13584738969802856, |
| "learning_rate": 4.845172067240415e-05, |
| "loss": 1.232811689376831, |
| "mean_token_accuracy": 0.6496494188904762, |
| "num_tokens": 4906642.0, |
| "step": 38 |
| }, |
| { |
| "entropy": 1.2297646403312683, |
| "epoch": 0.5016077170418006, |
| "grad_norm": 0.1445397287607193, |
| "learning_rate": 4.8326808849565936e-05, |
| "loss": 1.2179179191589355, |
| "mean_token_accuracy": 0.6524648442864418, |
| "num_tokens": 5035784.0, |
| "step": 39 |
| }, |
| { |
| "entropy": 1.2054394781589508, |
| "epoch": 0.5144694533762058, |
| "grad_norm": 0.1438300460577011, |
| "learning_rate": 4.819722568241274e-05, |
| "loss": 1.1902542114257812, |
| "mean_token_accuracy": 0.6586381196975708, |
| "num_tokens": 5163483.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.1801835894584656, |
| "epoch": 0.5273311897106109, |
| "grad_norm": 0.13536524772644043, |
| "learning_rate": 4.806299712081172e-05, |
| "loss": 1.1734623908996582, |
| "mean_token_accuracy": 0.6613527312874794, |
| "num_tokens": 5289099.0, |
| "step": 41 |
| }, |
| { |
| "entropy": 1.2093226611614227, |
| "epoch": 0.5401929260450161, |
| "grad_norm": 0.13810646533966064, |
| "learning_rate": 4.792415004490034e-05, |
| "loss": 1.2002174854278564, |
| "mean_token_accuracy": 0.6581337600946426, |
| "num_tokens": 5414735.0, |
| "step": 42 |
| }, |
| { |
| "entropy": 1.1848722696304321, |
| "epoch": 0.5530546623794212, |
| "grad_norm": 0.1468912959098816, |
| "learning_rate": 4.77807122597034e-05, |
| "loss": 1.1885192394256592, |
| "mean_token_accuracy": 0.6589618176221848, |
| "num_tokens": 5543143.0, |
| "step": 43 |
| }, |
| { |
| "entropy": 1.181750476360321, |
| "epoch": 0.5659163987138264, |
| "grad_norm": 0.1429019272327423, |
| "learning_rate": 4.7632712489564926e-05, |
| "loss": 1.1800310611724854, |
| "mean_token_accuracy": 0.6618404239416122, |
| "num_tokens": 5671758.0, |
| "step": 44 |
| }, |
| { |
| "entropy": 1.223798543214798, |
| "epoch": 0.5787781350482315, |
| "grad_norm": 0.13423405587673187, |
| "learning_rate": 4.748018037239592e-05, |
| "loss": 1.2137880325317383, |
| "mean_token_accuracy": 0.6538020968437195, |
| "num_tokens": 5801203.0, |
| "step": 45 |
| }, |
| { |
| "entropy": 1.1596673130989075, |
| "epoch": 0.5916398713826366, |
| "grad_norm": 0.14085975289344788, |
| "learning_rate": 4.732314645373921e-05, |
| "loss": 1.1383979320526123, |
| "mean_token_accuracy": 0.6676715686917305, |
| "num_tokens": 5929286.0, |
| "step": 46 |
| }, |
| { |
| "entropy": 1.1791094541549683, |
| "epoch": 0.6045016077170418, |
| "grad_norm": 0.14022590219974518, |
| "learning_rate": 4.7161642180652464e-05, |
| "loss": 1.1624398231506348, |
| "mean_token_accuracy": 0.6638200730085373, |
| "num_tokens": 6055494.0, |
| "step": 47 |
| }, |
| { |
| "entropy": 1.17191481590271, |
| "epoch": 0.617363344051447, |
| "grad_norm": 0.14044098556041718, |
| "learning_rate": 4.699569989541074e-05, |
| "loss": 1.1661159992218018, |
| "mean_token_accuracy": 0.6632223278284073, |
| "num_tokens": 6184519.0, |
| "step": 48 |
| }, |
| { |
| "entropy": 1.1261081546545029, |
| "epoch": 0.6302250803858521, |
| "grad_norm": 0.13071966171264648, |
| "learning_rate": 4.6825352829029705e-05, |
| "loss": 1.1280871629714966, |
| "mean_token_accuracy": 0.672090008854866, |
| "num_tokens": 6313996.0, |
| "step": 49 |
| }, |
| { |
| "entropy": 1.108861967921257, |
| "epoch": 0.6430868167202572, |
| "grad_norm": 0.13524308800697327, |
| "learning_rate": 4.665063509461097e-05, |
| "loss": 1.1100517511367798, |
| "mean_token_accuracy": 0.6758645325899124, |
| "num_tokens": 6442914.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.1622951924800873, |
| "epoch": 0.6559485530546624, |
| "grad_norm": 0.14350417256355286, |
| "learning_rate": 4.647158168051066e-05, |
| "loss": 1.1698243618011475, |
| "mean_token_accuracy": 0.661836288869381, |
| "num_tokens": 6569312.0, |
| "step": 51 |
| }, |
| { |
| "entropy": 1.1530523151159286, |
| "epoch": 0.6688102893890675, |
| "grad_norm": 0.13998498022556305, |
| "learning_rate": 4.628822844333278e-05, |
| "loss": 1.135171890258789, |
| "mean_token_accuracy": 0.6697984710335732, |
| "num_tokens": 6698149.0, |
| "step": 52 |
| }, |
| { |
| "entropy": 1.1557507067918777, |
| "epoch": 0.6816720257234726, |
| "grad_norm": 0.13826699554920197, |
| "learning_rate": 4.6100612100748765e-05, |
| "loss": 1.1333105564117432, |
| "mean_token_accuracy": 0.6696186140179634, |
| "num_tokens": 6827547.0, |
| "step": 53 |
| }, |
| { |
| "entropy": 1.1415271162986755, |
| "epoch": 0.6945337620578779, |
| "grad_norm": 0.14170631766319275, |
| "learning_rate": 4.59087702241444e-05, |
| "loss": 1.1154472827911377, |
| "mean_token_accuracy": 0.6728732585906982, |
| "num_tokens": 6956641.0, |
| "step": 54 |
| }, |
| { |
| "entropy": 1.133439689874649, |
| "epoch": 0.707395498392283, |
| "grad_norm": 0.19915136694908142, |
| "learning_rate": 4.571274123109606e-05, |
| "loss": 1.12148916721344, |
| "mean_token_accuracy": 0.6721195057034492, |
| "num_tokens": 7087072.0, |
| "step": 55 |
| }, |
| { |
| "entropy": 1.1475654691457748, |
| "epoch": 0.7202572347266881, |
| "grad_norm": 0.1380179524421692, |
| "learning_rate": 4.551256437767719e-05, |
| "loss": 1.147176742553711, |
| "mean_token_accuracy": 0.6669263690710068, |
| "num_tokens": 7217233.0, |
| "step": 56 |
| }, |
| { |
| "entropy": 1.1235886365175247, |
| "epoch": 0.7331189710610932, |
| "grad_norm": 0.14977428317070007, |
| "learning_rate": 4.530827975059715e-05, |
| "loss": 1.1248936653137207, |
| "mean_token_accuracy": 0.6717223599553108, |
| "num_tokens": 7345601.0, |
| "step": 57 |
| }, |
| { |
| "entropy": 1.0966329723596573, |
| "epoch": 0.7459807073954984, |
| "grad_norm": 0.13217344880104065, |
| "learning_rate": 4.5099928259173516e-05, |
| "loss": 1.0947030782699585, |
| "mean_token_accuracy": 0.6786246225237846, |
| "num_tokens": 7473430.0, |
| "step": 58 |
| }, |
| { |
| "entropy": 1.1365204900503159, |
| "epoch": 0.7588424437299035, |
| "grad_norm": 0.12866446375846863, |
| "learning_rate": 4.488755162713975e-05, |
| "loss": 1.126430869102478, |
| "mean_token_accuracy": 0.6728999614715576, |
| "num_tokens": 7602826.0, |
| "step": 59 |
| }, |
| { |
| "entropy": 1.1535387188196182, |
| "epoch": 0.7717041800643086, |
| "grad_norm": 0.1415134221315384, |
| "learning_rate": 4.467119238428975e-05, |
| "loss": 1.1418571472167969, |
| "mean_token_accuracy": 0.6701028272509575, |
| "num_tokens": 7732028.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.1234214305877686, |
| "epoch": 0.7845659163987139, |
| "grad_norm": 0.14994795620441437, |
| "learning_rate": 4.445089385796099e-05, |
| "loss": 1.1106842756271362, |
| "mean_token_accuracy": 0.6764309927821159, |
| "num_tokens": 7860385.0, |
| "step": 61 |
| }, |
| { |
| "entropy": 1.0901593267917633, |
| "epoch": 0.797427652733119, |
| "grad_norm": 0.13637474179267883, |
| "learning_rate": 4.422670016435792e-05, |
| "loss": 1.0811570882797241, |
| "mean_token_accuracy": 0.6831180602312088, |
| "num_tokens": 7989531.0, |
| "step": 62 |
| }, |
| { |
| "entropy": 1.1060381978750229, |
| "epoch": 0.8102893890675241, |
| "grad_norm": 0.1306847780942917, |
| "learning_rate": 4.3998656199717435e-05, |
| "loss": 1.1069831848144531, |
| "mean_token_accuracy": 0.6745292320847511, |
| "num_tokens": 8118824.0, |
| "step": 63 |
| }, |
| { |
| "entropy": 1.1469190865755081, |
| "epoch": 0.8231511254019293, |
| "grad_norm": 0.14836829900741577, |
| "learning_rate": 4.3766807631318106e-05, |
| "loss": 1.1478967666625977, |
| "mean_token_accuracy": 0.6674168556928635, |
| "num_tokens": 8248122.0, |
| "step": 64 |
| }, |
| { |
| "entropy": 1.1051507145166397, |
| "epoch": 0.8360128617363344, |
| "grad_norm": 0.1413162350654602, |
| "learning_rate": 4.353120088833501e-05, |
| "loss": 1.0933685302734375, |
| "mean_token_accuracy": 0.6783376038074493, |
| "num_tokens": 8376429.0, |
| "step": 65 |
| }, |
| { |
| "entropy": 1.1381653249263763, |
| "epoch": 0.8488745980707395, |
| "grad_norm": 0.13279159367084503, |
| "learning_rate": 4.329188315254196e-05, |
| "loss": 1.1261053085327148, |
| "mean_token_accuracy": 0.6716507449746132, |
| "num_tokens": 8504687.0, |
| "step": 66 |
| }, |
| { |
| "entropy": 1.1401405185461044, |
| "epoch": 0.8617363344051447, |
| "grad_norm": 0.14999179542064667, |
| "learning_rate": 4.3048902348863116e-05, |
| "loss": 1.122992992401123, |
| "mean_token_accuracy": 0.6721775308251381, |
| "num_tokens": 8635119.0, |
| "step": 67 |
| }, |
| { |
| "entropy": 1.1104646772146225, |
| "epoch": 0.8745980707395499, |
| "grad_norm": 0.15244296193122864, |
| "learning_rate": 4.280230713577564e-05, |
| "loss": 1.0963469743728638, |
| "mean_token_accuracy": 0.6784974113106728, |
| "num_tokens": 8765133.0, |
| "step": 68 |
| }, |
| { |
| "entropy": 1.1325141787528992, |
| "epoch": 0.887459807073955, |
| "grad_norm": 0.13338243961334229, |
| "learning_rate": 4.255214689556557e-05, |
| "loss": 1.133899211883545, |
| "mean_token_accuracy": 0.6714881733059883, |
| "num_tokens": 8895437.0, |
| "step": 69 |
| }, |
| { |
| "entropy": 1.1021241694688797, |
| "epoch": 0.9003215434083601, |
| "grad_norm": 0.14672274887561798, |
| "learning_rate": 4.229847172443866e-05, |
| "loss": 1.1050517559051514, |
| "mean_token_accuracy": 0.6782850474119186, |
| "num_tokens": 9023472.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.0665733218193054, |
| "epoch": 0.9131832797427653, |
| "grad_norm": 0.1326168030500412, |
| "learning_rate": 4.204133242248832e-05, |
| "loss": 1.0614970922470093, |
| "mean_token_accuracy": 0.6890874728560448, |
| "num_tokens": 9152924.0, |
| "step": 71 |
| }, |
| { |
| "entropy": 1.0804976969957352, |
| "epoch": 0.9260450160771704, |
| "grad_norm": 0.14258776605129242, |
| "learning_rate": 4.1780780483522575e-05, |
| "loss": 1.071770191192627, |
| "mean_token_accuracy": 0.6831802353262901, |
| "num_tokens": 9282930.0, |
| "step": 72 |
| }, |
| { |
| "entropy": 1.0698173642158508, |
| "epoch": 0.9389067524115756, |
| "grad_norm": 0.13697165250778198, |
| "learning_rate": 4.151686808475204e-05, |
| "loss": 1.057246208190918, |
| "mean_token_accuracy": 0.6859329044818878, |
| "num_tokens": 9410776.0, |
| "step": 73 |
| }, |
| { |
| "entropy": 1.0982066690921783, |
| "epoch": 0.9517684887459807, |
| "grad_norm": 0.14332003891468048, |
| "learning_rate": 4.1249648076341165e-05, |
| "loss": 1.08486008644104, |
| "mean_token_accuracy": 0.6810793280601501, |
| "num_tokens": 9539314.0, |
| "step": 74 |
| }, |
| { |
| "entropy": 1.1045535802841187, |
| "epoch": 0.9646302250803859, |
| "grad_norm": 0.14386016130447388, |
| "learning_rate": 4.0979173970824626e-05, |
| "loss": 1.1035948991775513, |
| "mean_token_accuracy": 0.6733438894152641, |
| "num_tokens": 9668102.0, |
| "step": 75 |
| }, |
| { |
| "entropy": 1.0924865305423737, |
| "epoch": 0.977491961414791, |
| "grad_norm": 0.13745297491550446, |
| "learning_rate": 4.070549993239106e-05, |
| "loss": 1.090521216392517, |
| "mean_token_accuracy": 0.6789770871400833, |
| "num_tokens": 9797707.0, |
| "step": 76 |
| }, |
| { |
| "entropy": 1.0984582901000977, |
| "epoch": 0.9903536977491961, |
| "grad_norm": 0.14490951597690582, |
| "learning_rate": 4.0428680766036384e-05, |
| "loss": 1.1076843738555908, |
| "mean_token_accuracy": 0.6759174689650536, |
| "num_tokens": 9927684.0, |
| "step": 77 |
| }, |
| { |
| "entropy": 1.0977885921796162, |
| "epoch": 1.0, |
| "grad_norm": 0.15585829317569733, |
| "learning_rate": 4.0148771906588706e-05, |
| "loss": 1.0940134525299072, |
| "mean_token_accuracy": 0.6789492865403494, |
| "num_tokens": 9998574.0, |
| "step": 78 |
| }, |
| { |
| "entropy": 1.1048902124166489, |
| "epoch": 1.0128617363344052, |
| "grad_norm": 0.13507987558841705, |
| "learning_rate": 3.986582940760717e-05, |
| "loss": 1.0746995210647583, |
| "mean_token_accuracy": 0.6833242624998093, |
| "num_tokens": 10127337.0, |
| "step": 79 |
| }, |
| { |
| "entropy": 1.1200366914272308, |
| "epoch": 1.0257234726688103, |
| "grad_norm": 0.14873428642749786, |
| "learning_rate": 3.957990993015683e-05, |
| "loss": 1.0768001079559326, |
| "mean_token_accuracy": 0.6837869137525558, |
| "num_tokens": 10253323.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.1033698171377182, |
| "epoch": 1.0385852090032155, |
| "grad_norm": 0.1395098865032196, |
| "learning_rate": 3.929107073146197e-05, |
| "loss": 1.0733070373535156, |
| "mean_token_accuracy": 0.6825986951589584, |
| "num_tokens": 10383034.0, |
| "step": 81 |
| }, |
| { |
| "entropy": 1.0916118025779724, |
| "epoch": 1.0514469453376205, |
| "grad_norm": 0.156707301735878, |
| "learning_rate": 3.899936965343989e-05, |
| "loss": 1.0789391994476318, |
| "mean_token_accuracy": 0.6803915277123451, |
| "num_tokens": 10512684.0, |
| "step": 82 |
| }, |
| { |
| "entropy": 1.0437547490000725, |
| "epoch": 1.0643086816720257, |
| "grad_norm": 0.14007939398288727, |
| "learning_rate": 3.8704865111117746e-05, |
| "loss": 1.046311378479004, |
| "mean_token_accuracy": 0.6876395344734192, |
| "num_tokens": 10642582.0, |
| "step": 83 |
| }, |
| { |
| "entropy": 1.0586023032665253, |
| "epoch": 1.077170418006431, |
| "grad_norm": 0.14878074824810028, |
| "learning_rate": 3.840761608093456e-05, |
| "loss": 1.0626013278961182, |
| "mean_token_accuracy": 0.6854664832353592, |
| "num_tokens": 10772130.0, |
| "step": 84 |
| }, |
| { |
| "entropy": 1.042947493493557, |
| "epoch": 1.090032154340836, |
| "grad_norm": 0.14047543704509735, |
| "learning_rate": 3.8107682088930794e-05, |
| "loss": 1.0462985038757324, |
| "mean_token_accuracy": 0.6900991648435593, |
| "num_tokens": 10902329.0, |
| "step": 85 |
| }, |
| { |
| "entropy": 1.0503267794847488, |
| "epoch": 1.1028938906752412, |
| "grad_norm": 0.13753578066825867, |
| "learning_rate": 3.7805123198827857e-05, |
| "loss": 1.0447890758514404, |
| "mean_token_accuracy": 0.6877391710877419, |
| "num_tokens": 11032146.0, |
| "step": 86 |
| }, |
| { |
| "entropy": 1.0541549250483513, |
| "epoch": 1.1157556270096463, |
| "grad_norm": 0.14761823415756226, |
| "learning_rate": 3.7500000000000003e-05, |
| "loss": 1.0494050979614258, |
| "mean_token_accuracy": 0.6881443187594414, |
| "num_tokens": 11162722.0, |
| "step": 87 |
| }, |
| { |
| "entropy": 1.0610276013612747, |
| "epoch": 1.1286173633440515, |
| "grad_norm": 0.13982513546943665, |
| "learning_rate": 3.719237359534087e-05, |
| "loss": 1.0486375093460083, |
| "mean_token_accuracy": 0.6868740543723106, |
| "num_tokens": 11291557.0, |
| "step": 88 |
| }, |
| { |
| "entropy": 1.0592399910092354, |
| "epoch": 1.1414790996784565, |
| "grad_norm": 0.14512217044830322, |
| "learning_rate": 3.688230558902725e-05, |
| "loss": 1.0511451959609985, |
| "mean_token_accuracy": 0.6883272528648376, |
| "num_tokens": 11420683.0, |
| "step": 89 |
| }, |
| { |
| "entropy": 1.1003997921943665, |
| "epoch": 1.1543408360128617, |
| "grad_norm": 0.13833674788475037, |
| "learning_rate": 3.656985807418248e-05, |
| "loss": 1.0804483890533447, |
| "mean_token_accuracy": 0.6791831701993942, |
| "num_tokens": 11549517.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.058895729482174, |
| "epoch": 1.167202572347267, |
| "grad_norm": 0.14019837975502014, |
| "learning_rate": 3.6255093620441834e-05, |
| "loss": 1.0490373373031616, |
| "mean_token_accuracy": 0.6885469257831573, |
| "num_tokens": 11678919.0, |
| "step": 91 |
| }, |
| { |
| "entropy": 1.0490736663341522, |
| "epoch": 1.180064308681672, |
| "grad_norm": 0.15101630985736847, |
| "learning_rate": 3.593807526142261e-05, |
| "loss": 1.0475131273269653, |
| "mean_token_accuracy": 0.6890112012624741, |
| "num_tokens": 11808171.0, |
| "step": 92 |
| }, |
| { |
| "entropy": 1.067186251282692, |
| "epoch": 1.1929260450160772, |
| "grad_norm": 0.14192728698253632, |
| "learning_rate": 3.56188664821012e-05, |
| "loss": 1.068861961364746, |
| "mean_token_accuracy": 0.6839301511645317, |
| "num_tokens": 11937495.0, |
| "step": 93 |
| }, |
| { |
| "entropy": 1.0729520842432976, |
| "epoch": 1.2057877813504823, |
| "grad_norm": 0.13931463658809662, |
| "learning_rate": 3.529753120609982e-05, |
| "loss": 1.0626987218856812, |
| "mean_token_accuracy": 0.6865019202232361, |
| "num_tokens": 12065721.0, |
| "step": 94 |
| }, |
| { |
| "entropy": 1.0651484578847885, |
| "epoch": 1.2186495176848875, |
| "grad_norm": 0.14170296490192413, |
| "learning_rate": 3.497413378288541e-05, |
| "loss": 1.065182089805603, |
| "mean_token_accuracy": 0.6859815120697021, |
| "num_tokens": 12195222.0, |
| "step": 95 |
| }, |
| { |
| "entropy": 1.0627811029553413, |
| "epoch": 1.2315112540192925, |
| "grad_norm": 0.15864631533622742, |
| "learning_rate": 3.464873897488322e-05, |
| "loss": 1.064727783203125, |
| "mean_token_accuracy": 0.6850855350494385, |
| "num_tokens": 12325399.0, |
| "step": 96 |
| }, |
| { |
| "entropy": 1.0540687665343285, |
| "epoch": 1.2443729903536977, |
| "grad_norm": 0.1563236117362976, |
| "learning_rate": 3.432141194450772e-05, |
| "loss": 1.0473036766052246, |
| "mean_token_accuracy": 0.6896297782659531, |
| "num_tokens": 12455203.0, |
| "step": 97 |
| }, |
| { |
| "entropy": 1.0381047055125237, |
| "epoch": 1.257234726688103, |
| "grad_norm": 0.1378972828388214, |
| "learning_rate": 3.39922182411134e-05, |
| "loss": 1.0351678133010864, |
| "mean_token_accuracy": 0.6909719184041023, |
| "num_tokens": 12582864.0, |
| "step": 98 |
| }, |
| { |
| "entropy": 1.0567686855793, |
| "epoch": 1.270096463022508, |
| "grad_norm": 0.1447557955980301, |
| "learning_rate": 3.3661223787868094e-05, |
| "loss": 1.045164704322815, |
| "mean_token_accuracy": 0.6886090338230133, |
| "num_tokens": 12711992.0, |
| "step": 99 |
| }, |
| { |
| "entropy": 1.0664659440517426, |
| "epoch": 1.2829581993569132, |
| "grad_norm": 0.14579758048057556, |
| "learning_rate": 3.332849486855144e-05, |
| "loss": 1.0578923225402832, |
| "mean_token_accuracy": 0.6857578605413437, |
| "num_tokens": 12841160.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.0501502603292465, |
| "epoch": 1.2958199356913183, |
| "grad_norm": 0.14204849302768707, |
| "learning_rate": 3.2994098114281134e-05, |
| "loss": 1.0381393432617188, |
| "mean_token_accuracy": 0.6887901425361633, |
| "num_tokens": 12967285.0, |
| "step": 101 |
| }, |
| { |
| "entropy": 1.0534097403287888, |
| "epoch": 1.3086816720257235, |
| "grad_norm": 0.1419542282819748, |
| "learning_rate": 3.265810049016959e-05, |
| "loss": 1.0446325540542603, |
| "mean_token_accuracy": 0.6880638003349304, |
| "num_tokens": 13095177.0, |
| "step": 102 |
| }, |
| { |
| "entropy": 1.0461725443601608, |
| "epoch": 1.3215434083601285, |
| "grad_norm": 0.14444847404956818, |
| "learning_rate": 3.232056928191376e-05, |
| "loss": 1.032381296157837, |
| "mean_token_accuracy": 0.691328376531601, |
| "num_tokens": 13224870.0, |
| "step": 103 |
| }, |
| { |
| "entropy": 1.0287123173475266, |
| "epoch": 1.3344051446945338, |
| "grad_norm": 0.14262095093727112, |
| "learning_rate": 3.1981572082320756e-05, |
| "loss": 1.0161638259887695, |
| "mean_token_accuracy": 0.6958137080073357, |
| "num_tokens": 13355092.0, |
| "step": 104 |
| }, |
| { |
| "entropy": 1.0425420626997948, |
| "epoch": 1.347266881028939, |
| "grad_norm": 0.14054642617702484, |
| "learning_rate": 3.164117677777191e-05, |
| "loss": 1.037819266319275, |
| "mean_token_accuracy": 0.690584309399128, |
| "num_tokens": 13482910.0, |
| "step": 105 |
| }, |
| { |
| "entropy": 1.0867420881986618, |
| "epoch": 1.360128617363344, |
| "grad_norm": 0.14352700114250183, |
| "learning_rate": 3.1299451534628135e-05, |
| "loss": 1.0806154012680054, |
| "mean_token_accuracy": 0.6801560521125793, |
| "num_tokens": 13610860.0, |
| "step": 106 |
| }, |
| { |
| "entropy": 1.0425082966685295, |
| "epoch": 1.3729903536977492, |
| "grad_norm": 0.14772510528564453, |
| "learning_rate": 3.0956464785579124e-05, |
| "loss": 1.0282400846481323, |
| "mean_token_accuracy": 0.6922867074608803, |
| "num_tokens": 13739967.0, |
| "step": 107 |
| }, |
| { |
| "entropy": 1.033432051539421, |
| "epoch": 1.3858520900321543, |
| "grad_norm": 0.1441124826669693, |
| "learning_rate": 3.061228521593931e-05, |
| "loss": 1.0348222255706787, |
| "mean_token_accuracy": 0.6906738951802254, |
| "num_tokens": 13869092.0, |
| "step": 108 |
| }, |
| { |
| "entropy": 1.0250284597277641, |
| "epoch": 1.3987138263665595, |
| "grad_norm": 0.14241281151771545, |
| "learning_rate": 3.0266981749893157e-05, |
| "loss": 1.019615650177002, |
| "mean_token_accuracy": 0.6946801096200943, |
| "num_tokens": 13999751.0, |
| "step": 109 |
| }, |
| { |
| "entropy": 1.0539506748318672, |
| "epoch": 1.4115755627009645, |
| "grad_norm": 0.14640918374061584, |
| "learning_rate": 2.9920623536692638e-05, |
| "loss": 1.0554652214050293, |
| "mean_token_accuracy": 0.6877814680337906, |
| "num_tokens": 14126193.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.0477852746844292, |
| "epoch": 1.4244372990353698, |
| "grad_norm": 0.13899168372154236, |
| "learning_rate": 2.9573279936809667e-05, |
| "loss": 1.0390174388885498, |
| "mean_token_accuracy": 0.6889165490865707, |
| "num_tokens": 14255192.0, |
| "step": 111 |
| }, |
| { |
| "entropy": 1.028913527727127, |
| "epoch": 1.437299035369775, |
| "grad_norm": 0.14846587181091309, |
| "learning_rate": 2.9225020508046232e-05, |
| "loss": 1.02781343460083, |
| "mean_token_accuracy": 0.6925171986222267, |
| "num_tokens": 14383678.0, |
| "step": 112 |
| }, |
| { |
| "entropy": 1.040368527173996, |
| "epoch": 1.45016077170418, |
| "grad_norm": 0.14321881532669067, |
| "learning_rate": 2.8875914991604948e-05, |
| "loss": 1.040635108947754, |
| "mean_token_accuracy": 0.6885931342840195, |
| "num_tokens": 14513886.0, |
| "step": 113 |
| }, |
| { |
| "entropy": 1.0307907313108444, |
| "epoch": 1.4630225080385852, |
| "grad_norm": 0.13824157416820526, |
| "learning_rate": 2.8526033298122985e-05, |
| "loss": 1.0185027122497559, |
| "mean_token_accuracy": 0.6952621415257454, |
| "num_tokens": 14644185.0, |
| "step": 114 |
| }, |
| { |
| "entropy": 1.0155034512281418, |
| "epoch": 1.4758842443729905, |
| "grad_norm": 0.14209888875484467, |
| "learning_rate": 2.8175445493671972e-05, |
| "loss": 0.996734619140625, |
| "mean_token_accuracy": 0.6990666836500168, |
| "num_tokens": 14769497.0, |
| "step": 115 |
| }, |
| { |
| "entropy": 1.021187648177147, |
| "epoch": 1.4887459807073955, |
| "grad_norm": 0.14187516272068024, |
| "learning_rate": 2.782422178572682e-05, |
| "loss": 1.0097088813781738, |
| "mean_token_accuracy": 0.6962587013840675, |
| "num_tokens": 14898899.0, |
| "step": 116 |
| }, |
| { |
| "entropy": 1.0390162095427513, |
| "epoch": 1.5016077170418005, |
| "grad_norm": 0.1539003998041153, |
| "learning_rate": 2.7472432509106248e-05, |
| "loss": 1.0348460674285889, |
| "mean_token_accuracy": 0.6902582123875618, |
| "num_tokens": 15029008.0, |
| "step": 117 |
| }, |
| { |
| "entropy": 1.084839791059494, |
| "epoch": 1.5144694533762058, |
| "grad_norm": 0.15639474987983704, |
| "learning_rate": 2.7120148111887732e-05, |
| "loss": 1.0750864744186401, |
| "mean_token_accuracy": 0.6826967149972916, |
| "num_tokens": 15157238.0, |
| "step": 118 |
| }, |
| { |
| "entropy": 1.0497614741325378, |
| "epoch": 1.527331189710611, |
| "grad_norm": 0.14684948325157166, |
| "learning_rate": 2.6767439141299865e-05, |
| "loss": 1.0455033779144287, |
| "mean_token_accuracy": 0.6880874261260033, |
| "num_tokens": 15285438.0, |
| "step": 119 |
| }, |
| { |
| "entropy": 1.03941510617733, |
| "epoch": 1.540192926045016, |
| "grad_norm": 0.143926739692688, |
| "learning_rate": 2.6414376229594813e-05, |
| "loss": 1.0348516702651978, |
| "mean_token_accuracy": 0.6904465183615685, |
| "num_tokens": 15413195.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.0160878226161003, |
| "epoch": 1.5530546623794212, |
| "grad_norm": 0.14575257897377014, |
| "learning_rate": 2.606103007990371e-05, |
| "loss": 1.006327509880066, |
| "mean_token_accuracy": 0.6950085982680321, |
| "num_tokens": 15542436.0, |
| "step": 121 |
| }, |
| { |
| "entropy": 1.054566353559494, |
| "epoch": 1.5659163987138265, |
| "grad_norm": 0.14473803341388702, |
| "learning_rate": 2.570747145207796e-05, |
| "loss": 1.0497708320617676, |
| "mean_token_accuracy": 0.6880283355712891, |
| "num_tokens": 15669820.0, |
| "step": 122 |
| }, |
| { |
| "entropy": 1.0369255393743515, |
| "epoch": 1.5787781350482315, |
| "grad_norm": 0.14907261729240417, |
| "learning_rate": 2.5353771148519057e-05, |
| "loss": 1.0358316898345947, |
| "mean_token_accuracy": 0.6919823661446571, |
| "num_tokens": 15797917.0, |
| "step": 123 |
| }, |
| { |
| "entropy": 1.029112845659256, |
| "epoch": 1.5916398713826365, |
| "grad_norm": 0.14412133395671844, |
| "learning_rate": 2.5e-05, |
| "loss": 1.0267410278320312, |
| "mean_token_accuracy": 0.6918147578835487, |
| "num_tokens": 15926059.0, |
| "step": 124 |
| }, |
| { |
| "entropy": 1.0285268500447273, |
| "epoch": 1.6045016077170418, |
| "grad_norm": 0.1455468088388443, |
| "learning_rate": 2.4646228851480956e-05, |
| "loss": 1.0172383785247803, |
| "mean_token_accuracy": 0.6961864978075027, |
| "num_tokens": 16055416.0, |
| "step": 125 |
| }, |
| { |
| "entropy": 1.0407991334795952, |
| "epoch": 1.617363344051447, |
| "grad_norm": 0.14318124949932098, |
| "learning_rate": 2.429252854792205e-05, |
| "loss": 1.0395405292510986, |
| "mean_token_accuracy": 0.6874824613332748, |
| "num_tokens": 16185719.0, |
| "step": 126 |
| }, |
| { |
| "entropy": 1.0249589011073112, |
| "epoch": 1.630225080385852, |
| "grad_norm": 0.1448572725057602, |
| "learning_rate": 2.39389699200963e-05, |
| "loss": 1.015451431274414, |
| "mean_token_accuracy": 0.6942317560315132, |
| "num_tokens": 16314607.0, |
| "step": 127 |
| }, |
| { |
| "entropy": 1.039283350110054, |
| "epoch": 1.6430868167202572, |
| "grad_norm": 0.1484907567501068, |
| "learning_rate": 2.358562377040519e-05, |
| "loss": 1.028055191040039, |
| "mean_token_accuracy": 0.6915712878108025, |
| "num_tokens": 16441937.0, |
| "step": 128 |
| }, |
| { |
| "entropy": 1.0223650634288788, |
| "epoch": 1.6559485530546625, |
| "grad_norm": 0.14289766550064087, |
| "learning_rate": 2.3232560858700137e-05, |
| "loss": 1.008897304534912, |
| "mean_token_accuracy": 0.6952646598219872, |
| "num_tokens": 16571389.0, |
| "step": 129 |
| }, |
| { |
| "entropy": 0.9995019808411598, |
| "epoch": 1.6688102893890675, |
| "grad_norm": 0.14972709119319916, |
| "learning_rate": 2.287985188811228e-05, |
| "loss": 0.987511157989502, |
| "mean_token_accuracy": 0.7003577873110771, |
| "num_tokens": 16701314.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.0168959125876427, |
| "epoch": 1.6816720257234725, |
| "grad_norm": 0.1496725082397461, |
| "learning_rate": 2.2527567490893758e-05, |
| "loss": 1.0123668909072876, |
| "mean_token_accuracy": 0.6951291859149933, |
| "num_tokens": 16830317.0, |
| "step": 131 |
| }, |
| { |
| "entropy": 1.0190805196762085, |
| "epoch": 1.694533762057878, |
| "grad_norm": 0.15911011397838593, |
| "learning_rate": 2.2175778214273185e-05, |
| "loss": 1.0175803899765015, |
| "mean_token_accuracy": 0.6933339908719063, |
| "num_tokens": 16959965.0, |
| "step": 132 |
| }, |
| { |
| "entropy": 1.0380867347121239, |
| "epoch": 1.707395498392283, |
| "grad_norm": 0.14191225171089172, |
| "learning_rate": 2.182455450632803e-05, |
| "loss": 1.0294275283813477, |
| "mean_token_accuracy": 0.6917987540364265, |
| "num_tokens": 17089870.0, |
| "step": 133 |
| }, |
| { |
| "entropy": 1.0180665403604507, |
| "epoch": 1.720257234726688, |
| "grad_norm": 0.1482504904270172, |
| "learning_rate": 2.1473966701877025e-05, |
| "loss": 1.0076394081115723, |
| "mean_token_accuracy": 0.6966089159250259, |
| "num_tokens": 17218254.0, |
| "step": 134 |
| }, |
| { |
| "entropy": 1.0091007053852081, |
| "epoch": 1.7331189710610932, |
| "grad_norm": 0.15382039546966553, |
| "learning_rate": 2.1124085008395054e-05, |
| "loss": 1.003111481666565, |
| "mean_token_accuracy": 0.70043034106493, |
| "num_tokens": 17347960.0, |
| "step": 135 |
| }, |
| { |
| "entropy": 1.0102703422307968, |
| "epoch": 1.7459807073954985, |
| "grad_norm": 0.18475782871246338, |
| "learning_rate": 2.0774979491953777e-05, |
| "loss": 1.0030856132507324, |
| "mean_token_accuracy": 0.6967976689338684, |
| "num_tokens": 17476832.0, |
| "step": 136 |
| }, |
| { |
| "entropy": 1.0280822068452835, |
| "epoch": 1.7588424437299035, |
| "grad_norm": 0.15235531330108643, |
| "learning_rate": 2.0426720063190335e-05, |
| "loss": 1.0269193649291992, |
| "mean_token_accuracy": 0.6915133595466614, |
| "num_tokens": 17606491.0, |
| "step": 137 |
| }, |
| { |
| "entropy": 1.038973644375801, |
| "epoch": 1.7717041800643085, |
| "grad_norm": 0.1467663049697876, |
| "learning_rate": 2.0079376463307368e-05, |
| "loss": 1.0385396480560303, |
| "mean_token_accuracy": 0.6904850155115128, |
| "num_tokens": 17735828.0, |
| "step": 138 |
| }, |
| { |
| "entropy": 1.007546715438366, |
| "epoch": 1.784565916398714, |
| "grad_norm": 0.15167458355426788, |
| "learning_rate": 1.973301825010685e-05, |
| "loss": 1.0009288787841797, |
| "mean_token_accuracy": 0.6996477469801903, |
| "num_tokens": 17865343.0, |
| "step": 139 |
| }, |
| { |
| "entropy": 1.0457243248820305, |
| "epoch": 1.797427652733119, |
| "grad_norm": 0.14147217571735382, |
| "learning_rate": 1.9387714784060685e-05, |
| "loss": 1.0354124307632446, |
| "mean_token_accuracy": 0.6904323399066925, |
| "num_tokens": 17993892.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.0493519008159637, |
| "epoch": 1.810289389067524, |
| "grad_norm": 0.14349812269210815, |
| "learning_rate": 1.904353521442088e-05, |
| "loss": 1.043258786201477, |
| "mean_token_accuracy": 0.6896533966064453, |
| "num_tokens": 18122215.0, |
| "step": 141 |
| }, |
| { |
| "entropy": 1.0361140668392181, |
| "epoch": 1.8231511254019293, |
| "grad_norm": 0.14594632387161255, |
| "learning_rate": 1.8700548465371874e-05, |
| "loss": 1.0246191024780273, |
| "mean_token_accuracy": 0.6924104690551758, |
| "num_tokens": 18250505.0, |
| "step": 142 |
| }, |
| { |
| "entropy": 0.996397003531456, |
| "epoch": 1.8360128617363345, |
| "grad_norm": 0.13873013854026794, |
| "learning_rate": 1.8358823222228097e-05, |
| "loss": 0.9915734529495239, |
| "mean_token_accuracy": 0.6993561685085297, |
| "num_tokens": 18381218.0, |
| "step": 143 |
| }, |
| { |
| "entropy": 1.0313754975795746, |
| "epoch": 1.8488745980707395, |
| "grad_norm": 0.15026451647281647, |
| "learning_rate": 1.801842791767925e-05, |
| "loss": 1.016987681388855, |
| "mean_token_accuracy": 0.6936725974082947, |
| "num_tokens": 18510172.0, |
| "step": 144 |
| }, |
| { |
| "entropy": 1.0371734127402306, |
| "epoch": 1.8617363344051447, |
| "grad_norm": 0.14328952133655548, |
| "learning_rate": 1.7679430718086243e-05, |
| "loss": 1.029707908630371, |
| "mean_token_accuracy": 0.6911671832203865, |
| "num_tokens": 18639469.0, |
| "step": 145 |
| }, |
| { |
| "entropy": 1.0529436096549034, |
| "epoch": 1.87459807073955, |
| "grad_norm": 0.15134425461292267, |
| "learning_rate": 1.7341899509830416e-05, |
| "loss": 1.0452293157577515, |
| "mean_token_accuracy": 0.6891046240925789, |
| "num_tokens": 18769333.0, |
| "step": 146 |
| }, |
| { |
| "entropy": 1.0049605518579483, |
| "epoch": 1.887459807073955, |
| "grad_norm": 0.1447063535451889, |
| "learning_rate": 1.700590188571887e-05, |
| "loss": 1.0040454864501953, |
| "mean_token_accuracy": 0.6978042870759964, |
| "num_tokens": 18897391.0, |
| "step": 147 |
| }, |
| { |
| "entropy": 1.0140637308359146, |
| "epoch": 1.90032154340836, |
| "grad_norm": 0.1438555121421814, |
| "learning_rate": 1.667150513144856e-05, |
| "loss": 1.007043480873108, |
| "mean_token_accuracy": 0.698374018073082, |
| "num_tokens": 19026977.0, |
| "step": 148 |
| }, |
| { |
| "entropy": 1.0184678137302399, |
| "epoch": 1.9131832797427653, |
| "grad_norm": 0.1421106904745102, |
| "learning_rate": 1.633877621213192e-05, |
| "loss": 1.01150381565094, |
| "mean_token_accuracy": 0.6960654258728027, |
| "num_tokens": 19154514.0, |
| "step": 149 |
| }, |
| { |
| "entropy": 1.0100202932953835, |
| "epoch": 1.9260450160771705, |
| "grad_norm": 0.1437564641237259, |
| "learning_rate": 1.6007781758886607e-05, |
| "loss": 1.003399133682251, |
| "mean_token_accuracy": 0.6989860460162163, |
| "num_tokens": 19282293.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.0097703114151955, |
| "epoch": 1.9389067524115755, |
| "grad_norm": 0.14791278541088104, |
| "learning_rate": 1.567858805549229e-05, |
| "loss": 1.0031559467315674, |
| "mean_token_accuracy": 0.6982970088720322, |
| "num_tokens": 19412057.0, |
| "step": 151 |
| }, |
| { |
| "entropy": 0.9986948445439339, |
| "epoch": 1.9517684887459807, |
| "grad_norm": 0.14181585609912872, |
| "learning_rate": 1.535126102511678e-05, |
| "loss": 0.9911055564880371, |
| "mean_token_accuracy": 0.7003660649061203, |
| "num_tokens": 19538894.0, |
| "step": 152 |
| }, |
| { |
| "entropy": 1.0326199233531952, |
| "epoch": 1.964630225080386, |
| "grad_norm": 0.14459078013896942, |
| "learning_rate": 1.5025866217114592e-05, |
| "loss": 1.029758095741272, |
| "mean_token_accuracy": 0.690429113805294, |
| "num_tokens": 19669050.0, |
| "step": 153 |
| }, |
| { |
| "entropy": 1.024147979915142, |
| "epoch": 1.977491961414791, |
| "grad_norm": 0.14128583669662476, |
| "learning_rate": 1.4702468793900188e-05, |
| "loss": 1.0202356576919556, |
| "mean_token_accuracy": 0.6920891180634499, |
| "num_tokens": 19796666.0, |
| "step": 154 |
| }, |
| { |
| "entropy": 1.0354069098830223, |
| "epoch": 1.990353697749196, |
| "grad_norm": 0.14648661017417908, |
| "learning_rate": 1.4381133517898804e-05, |
| "loss": 1.0325945615768433, |
| "mean_token_accuracy": 0.6926636919379234, |
| "num_tokens": 19927210.0, |
| "step": 155 |
| }, |
| { |
| "entropy": 0.9808994829654694, |
| "epoch": 2.0, |
| "grad_norm": 0.1645454615354538, |
| "learning_rate": 1.4061924738577403e-05, |
| "loss": 0.992160975933075, |
| "mean_token_accuracy": 0.706730475028356, |
| "num_tokens": 19999816.0, |
| "step": 156 |
| }, |
| { |
| "entropy": 1.0143855810165405, |
| "epoch": 2.012861736334405, |
| "grad_norm": 0.15097208321094513, |
| "learning_rate": 1.3744906379558165e-05, |
| "loss": 0.9856299161911011, |
| "mean_token_accuracy": 0.7038262039422989, |
| "num_tokens": 20129551.0, |
| "step": 157 |
| }, |
| { |
| "entropy": 0.9946700483560562, |
| "epoch": 2.0257234726688105, |
| "grad_norm": 0.14270229637622833, |
| "learning_rate": 1.3430141925817532e-05, |
| "loss": 0.9719038605690002, |
| "mean_token_accuracy": 0.7039921209216118, |
| "num_tokens": 20259647.0, |
| "step": 158 |
| }, |
| { |
| "entropy": 1.0147498473525047, |
| "epoch": 2.0385852090032155, |
| "grad_norm": 0.14715144038200378, |
| "learning_rate": 1.3117694410972748e-05, |
| "loss": 0.9927316904067993, |
| "mean_token_accuracy": 0.698149062693119, |
| "num_tokens": 20389510.0, |
| "step": 159 |
| }, |
| { |
| "entropy": 1.0180071666836739, |
| "epoch": 2.0514469453376205, |
| "grad_norm": 0.14939357340335846, |
| "learning_rate": 1.2807626404659142e-05, |
| "loss": 0.9938769340515137, |
| "mean_token_accuracy": 0.6993245705962181, |
| "num_tokens": 20518012.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.0051729753613472, |
| "epoch": 2.0643086816720255, |
| "grad_norm": 0.14741602540016174, |
| "learning_rate": 1.2500000000000006e-05, |
| "loss": 0.9840617179870605, |
| "mean_token_accuracy": 0.7015254199504852, |
| "num_tokens": 20646832.0, |
| "step": 161 |
| }, |
| { |
| "entropy": 0.998705729842186, |
| "epoch": 2.077170418006431, |
| "grad_norm": 0.15145288407802582, |
| "learning_rate": 1.2194876801172148e-05, |
| "loss": 0.9809376001358032, |
| "mean_token_accuracy": 0.7015095204114914, |
| "num_tokens": 20774143.0, |
| "step": 162 |
| }, |
| { |
| "entropy": 1.0045012831687927, |
| "epoch": 2.090032154340836, |
| "grad_norm": 0.1488005369901657, |
| "learning_rate": 1.1892317911069212e-05, |
| "loss": 0.9931153059005737, |
| "mean_token_accuracy": 0.6996508315205574, |
| "num_tokens": 20900771.0, |
| "step": 163 |
| }, |
| { |
| "entropy": 0.9704049155116081, |
| "epoch": 2.102893890675241, |
| "grad_norm": 0.14846347272396088, |
| "learning_rate": 1.1592383919065442e-05, |
| "loss": 0.9699604511260986, |
| "mean_token_accuracy": 0.7040959894657135, |
| "num_tokens": 21030176.0, |
| "step": 164 |
| }, |
| { |
| "entropy": 0.9908866882324219, |
| "epoch": 2.1157556270096465, |
| "grad_norm": 0.15188701450824738, |
| "learning_rate": 1.1295134888882258e-05, |
| "loss": 0.9888017177581787, |
| "mean_token_accuracy": 0.7019331976771355, |
| "num_tokens": 21158984.0, |
| "step": 165 |
| }, |
| { |
| "entropy": 0.9876426234841347, |
| "epoch": 2.1286173633440515, |
| "grad_norm": 0.14991621673107147, |
| "learning_rate": 1.1000630346560117e-05, |
| "loss": 0.9930974245071411, |
| "mean_token_accuracy": 0.7000587806105614, |
| "num_tokens": 21287794.0, |
| "step": 166 |
| }, |
| { |
| "entropy": 0.9993386119604111, |
| "epoch": 2.1414790996784565, |
| "grad_norm": 0.15167978405952454, |
| "learning_rate": 1.0708929268538034e-05, |
| "loss": 1.0022180080413818, |
| "mean_token_accuracy": 0.6992425248026848, |
| "num_tokens": 21416220.0, |
| "step": 167 |
| }, |
| { |
| "entropy": 0.9697343483567238, |
| "epoch": 2.154340836012862, |
| "grad_norm": 0.15201398730278015, |
| "learning_rate": 1.0420090069843167e-05, |
| "loss": 0.9746421575546265, |
| "mean_token_accuracy": 0.7043134346604347, |
| "num_tokens": 21544581.0, |
| "step": 168 |
| }, |
| { |
| "entropy": 0.9662765711545944, |
| "epoch": 2.167202572347267, |
| "grad_norm": 0.14762073755264282, |
| "learning_rate": 1.0134170592392836e-05, |
| "loss": 0.9753881096839905, |
| "mean_token_accuracy": 0.702050969004631, |
| "num_tokens": 21674877.0, |
| "step": 169 |
| }, |
| { |
| "entropy": 0.9745886772871017, |
| "epoch": 2.180064308681672, |
| "grad_norm": 0.15288156270980835, |
| "learning_rate": 9.851228093411296e-06, |
| "loss": 0.9781433939933777, |
| "mean_token_accuracy": 0.7020246610045433, |
| "num_tokens": 21804650.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.9887658506631851, |
| "epoch": 2.192926045016077, |
| "grad_norm": 0.1461670845746994, |
| "learning_rate": 9.571319233963627e-06, |
| "loss": 0.9779214262962341, |
| "mean_token_accuracy": 0.703081913292408, |
| "num_tokens": 21933944.0, |
| "step": 171 |
| }, |
| { |
| "entropy": 0.972991868853569, |
| "epoch": 2.2057877813504825, |
| "grad_norm": 0.14857877790927887, |
| "learning_rate": 9.29450006760894e-06, |
| "loss": 0.9676029086112976, |
| "mean_token_accuracy": 0.7069084197282791, |
| "num_tokens": 22062831.0, |
| "step": 172 |
| }, |
| { |
| "entropy": 1.0113277956843376, |
| "epoch": 2.2186495176848875, |
| "grad_norm": 0.15487462282180786, |
| "learning_rate": 9.020826029175384e-06, |
| "loss": 1.006070613861084, |
| "mean_token_accuracy": 0.6965017765760422, |
| "num_tokens": 22191464.0, |
| "step": 173 |
| }, |
| { |
| "entropy": 0.9916297346353531, |
| "epoch": 2.2315112540192925, |
| "grad_norm": 0.14774714410305023, |
| "learning_rate": 8.750351923658832e-06, |
| "loss": 0.9754974842071533, |
| "mean_token_accuracy": 0.7031588032841682, |
| "num_tokens": 22321437.0, |
| "step": 174 |
| }, |
| { |
| "entropy": 1.0356281399726868, |
| "epoch": 2.244372990353698, |
| "grad_norm": 0.15343612432479858, |
| "learning_rate": 8.483131915247968e-06, |
| "loss": 1.017188549041748, |
| "mean_token_accuracy": 0.693478174507618, |
| "num_tokens": 22449634.0, |
| "step": 175 |
| }, |
| { |
| "entropy": 1.0122131705284119, |
| "epoch": 2.257234726688103, |
| "grad_norm": 0.14862875640392303, |
| "learning_rate": 8.219219516477428e-06, |
| "loss": 0.9908279180526733, |
| "mean_token_accuracy": 0.6987990885972977, |
| "num_tokens": 22577836.0, |
| "step": 176 |
| }, |
| { |
| "entropy": 1.0197054371237755, |
| "epoch": 2.270096463022508, |
| "grad_norm": 0.15119275450706482, |
| "learning_rate": 7.958667577511683e-06, |
| "loss": 0.9980366826057434, |
| "mean_token_accuracy": 0.6982897073030472, |
| "num_tokens": 22706085.0, |
| "step": 177 |
| }, |
| { |
| "entropy": 1.000802867114544, |
| "epoch": 2.282958199356913, |
| "grad_norm": 0.14366647601127625, |
| "learning_rate": 7.701528275561348e-06, |
| "loss": 0.9737733602523804, |
| "mean_token_accuracy": 0.7044953107833862, |
| "num_tokens": 22834748.0, |
| "step": 178 |
| }, |
| { |
| "entropy": 0.9863439574837685, |
| "epoch": 2.2958199356913185, |
| "grad_norm": 0.14263279736042023, |
| "learning_rate": 7.447853104434438e-06, |
| "loss": 0.9751866459846497, |
| "mean_token_accuracy": 0.7027714774012566, |
| "num_tokens": 22961691.0, |
| "step": 179 |
| }, |
| { |
| "entropy": 0.9606952220201492, |
| "epoch": 2.3086816720257235, |
| "grad_norm": 0.15060292184352875, |
| "learning_rate": 7.197692864224365e-06, |
| "loss": 0.9443936347961426, |
| "mean_token_accuracy": 0.7106149792671204, |
| "num_tokens": 23090767.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.9733371436595917, |
| "epoch": 2.3215434083601285, |
| "grad_norm": 0.14608748257160187, |
| "learning_rate": 6.951097651136889e-06, |
| "loss": 0.9657639265060425, |
| "mean_token_accuracy": 0.706763505935669, |
| "num_tokens": 23220713.0, |
| "step": 181 |
| }, |
| { |
| "entropy": 0.9712415337562561, |
| "epoch": 2.334405144694534, |
| "grad_norm": 0.1487581729888916, |
| "learning_rate": 6.708116847458043e-06, |
| "loss": 0.959854781627655, |
| "mean_token_accuracy": 0.7069366499781609, |
| "num_tokens": 23349717.0, |
| "step": 182 |
| }, |
| { |
| "entropy": 0.9718986302614212, |
| "epoch": 2.347266881028939, |
| "grad_norm": 0.14684326946735382, |
| "learning_rate": 6.468799111665003e-06, |
| "loss": 0.9676629900932312, |
| "mean_token_accuracy": 0.7048570662736893, |
| "num_tokens": 23478878.0, |
| "step": 183 |
| }, |
| { |
| "entropy": 0.9988474324345589, |
| "epoch": 2.360128617363344, |
| "grad_norm": 0.14926955103874207, |
| "learning_rate": 6.23319236868189e-06, |
| "loss": 0.9989900588989258, |
| "mean_token_accuracy": 0.6979110315442085, |
| "num_tokens": 23609398.0, |
| "step": 184 |
| }, |
| { |
| "entropy": 1.0037187412381172, |
| "epoch": 2.372990353697749, |
| "grad_norm": 0.15062850713729858, |
| "learning_rate": 6.001343800282569e-06, |
| "loss": 1.0000722408294678, |
| "mean_token_accuracy": 0.6993377730250359, |
| "num_tokens": 23738018.0, |
| "step": 185 |
| }, |
| { |
| "entropy": 0.9782311767339706, |
| "epoch": 2.3858520900321545, |
| "grad_norm": 0.14419959485530853, |
| "learning_rate": 5.7732998356420796e-06, |
| "loss": 0.9765899777412415, |
| "mean_token_accuracy": 0.702816791832447, |
| "num_tokens": 23867248.0, |
| "step": 186 |
| }, |
| { |
| "entropy": 0.9830808639526367, |
| "epoch": 2.3987138263665595, |
| "grad_norm": 0.1496494710445404, |
| "learning_rate": 5.549106142039018e-06, |
| "loss": 0.9824045896530151, |
| "mean_token_accuracy": 0.7036417871713638, |
| "num_tokens": 23990063.0, |
| "step": 187 |
| }, |
| { |
| "entropy": 1.0006206557154655, |
| "epoch": 2.4115755627009645, |
| "grad_norm": 0.14769846200942993, |
| "learning_rate": 5.328807615710246e-06, |
| "loss": 0.9914774894714355, |
| "mean_token_accuracy": 0.7003512755036354, |
| "num_tokens": 24117998.0, |
| "step": 188 |
| }, |
| { |
| "entropy": 0.9960543289780617, |
| "epoch": 2.42443729903537, |
| "grad_norm": 0.14919231832027435, |
| "learning_rate": 5.112448372860257e-06, |
| "loss": 0.9938825368881226, |
| "mean_token_accuracy": 0.7001635283231735, |
| "num_tokens": 24246395.0, |
| "step": 189 |
| }, |
| { |
| "entropy": 0.969782643020153, |
| "epoch": 2.437299035369775, |
| "grad_norm": 0.1437661498785019, |
| "learning_rate": 4.900071740826489e-06, |
| "loss": 0.9636502265930176, |
| "mean_token_accuracy": 0.7052842155098915, |
| "num_tokens": 24375083.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.9985753521323204, |
| "epoch": 2.45016077170418, |
| "grad_norm": 0.1478334367275238, |
| "learning_rate": 4.691720249402856e-06, |
| "loss": 0.9916254878044128, |
| "mean_token_accuracy": 0.7017218321561813, |
| "num_tokens": 24504249.0, |
| "step": 191 |
| }, |
| { |
| "entropy": 1.0174183622002602, |
| "epoch": 2.463022508038585, |
| "grad_norm": 0.1463613063097, |
| "learning_rate": 4.487435622322814e-06, |
| "loss": 1.0077807903289795, |
| "mean_token_accuracy": 0.6966395974159241, |
| "num_tokens": 24633736.0, |
| "step": 192 |
| }, |
| { |
| "entropy": 1.0191529616713524, |
| "epoch": 2.4758842443729905, |
| "grad_norm": 0.1539534330368042, |
| "learning_rate": 4.2872587689039484e-06, |
| "loss": 1.02093505859375, |
| "mean_token_accuracy": 0.6929882690310478, |
| "num_tokens": 24764322.0, |
| "step": 193 |
| }, |
| { |
| "entropy": 0.9834636375308037, |
| "epoch": 2.4887459807073955, |
| "grad_norm": 0.14389380812644958, |
| "learning_rate": 4.091229775855598e-06, |
| "loss": 0.9757540225982666, |
| "mean_token_accuracy": 0.7044370919466019, |
| "num_tokens": 24894140.0, |
| "step": 194 |
| }, |
| { |
| "entropy": 0.9720609411597252, |
| "epoch": 2.5016077170418005, |
| "grad_norm": 0.14187121391296387, |
| "learning_rate": 3.8993878992512415e-06, |
| "loss": 0.959560215473175, |
| "mean_token_accuracy": 0.7082379162311554, |
| "num_tokens": 25023082.0, |
| "step": 195 |
| }, |
| { |
| "entropy": 1.0143597051501274, |
| "epoch": 2.514469453376206, |
| "grad_norm": 0.1516728550195694, |
| "learning_rate": 3.711771556667218e-06, |
| "loss": 0.9982469081878662, |
| "mean_token_accuracy": 0.6964723244309425, |
| "num_tokens": 25152092.0, |
| "step": 196 |
| }, |
| { |
| "entropy": 0.9826101660728455, |
| "epoch": 2.527331189710611, |
| "grad_norm": 0.14548565447330475, |
| "learning_rate": 3.5284183194893488e-06, |
| "loss": 0.9657119512557983, |
| "mean_token_accuracy": 0.7059699892997742, |
| "num_tokens": 25281262.0, |
| "step": 197 |
| }, |
| { |
| "entropy": 0.988068088889122, |
| "epoch": 2.540192926045016, |
| "grad_norm": 0.14584684371948242, |
| "learning_rate": 3.3493649053890326e-06, |
| "loss": 0.974069356918335, |
| "mean_token_accuracy": 0.7047757878899574, |
| "num_tokens": 25409681.0, |
| "step": 198 |
| }, |
| { |
| "entropy": 0.9741279184818268, |
| "epoch": 2.553054662379421, |
| "grad_norm": 0.1395328938961029, |
| "learning_rate": 3.1746471709702964e-06, |
| "loss": 0.9705372452735901, |
| "mean_token_accuracy": 0.7048429250717163, |
| "num_tokens": 25538523.0, |
| "step": 199 |
| }, |
| { |
| "entropy": 0.9832991883158684, |
| "epoch": 2.5659163987138265, |
| "grad_norm": 0.16734182834625244, |
| "learning_rate": 3.0043001045892695e-06, |
| "loss": 0.970828652381897, |
| "mean_token_accuracy": 0.7034261524677277, |
| "num_tokens": 25666896.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.9968943223357201, |
| "epoch": 2.5787781350482315, |
| "grad_norm": 0.14664772152900696, |
| "learning_rate": 2.8383578193475315e-06, |
| "loss": 0.9886881113052368, |
| "mean_token_accuracy": 0.7005996480584145, |
| "num_tokens": 25795814.0, |
| "step": 201 |
| }, |
| { |
| "entropy": 0.9984467178583145, |
| "epoch": 2.5916398713826365, |
| "grad_norm": 0.14841625094413757, |
| "learning_rate": 2.676853546260791e-06, |
| "loss": 0.9910497069358826, |
| "mean_token_accuracy": 0.7002272084355354, |
| "num_tokens": 25925524.0, |
| "step": 202 |
| }, |
| { |
| "entropy": 0.9846226349473, |
| "epoch": 2.604501607717042, |
| "grad_norm": 0.14504557847976685, |
| "learning_rate": 2.5198196276040782e-06, |
| "loss": 0.9722692966461182, |
| "mean_token_accuracy": 0.7056987211108208, |
| "num_tokens": 26055184.0, |
| "step": 203 |
| }, |
| { |
| "entropy": 1.0102949291467667, |
| "epoch": 2.617363344051447, |
| "grad_norm": 0.14757543802261353, |
| "learning_rate": 2.3672875104350844e-06, |
| "loss": 0.9945117235183716, |
| "mean_token_accuracy": 0.698897622525692, |
| "num_tokens": 26184353.0, |
| "step": 204 |
| }, |
| { |
| "entropy": 0.9994086995720863, |
| "epoch": 2.630225080385852, |
| "grad_norm": 0.14748479425907135, |
| "learning_rate": 2.219287740296605e-06, |
| "loss": 0.9939561486244202, |
| "mean_token_accuracy": 0.6975871846079826, |
| "num_tokens": 26313066.0, |
| "step": 205 |
| }, |
| { |
| "entropy": 0.9975132793188095, |
| "epoch": 2.643086816720257, |
| "grad_norm": 0.1443839818239212, |
| "learning_rate": 2.075849955099668e-06, |
| "loss": 0.9907892942428589, |
| "mean_token_accuracy": 0.699777752161026, |
| "num_tokens": 26442692.0, |
| "step": 206 |
| }, |
| { |
| "entropy": 1.0035064443945885, |
| "epoch": 2.6559485530546625, |
| "grad_norm": 0.14282147586345673, |
| "learning_rate": 1.937002879188285e-06, |
| "loss": 0.9950806498527527, |
| "mean_token_accuracy": 0.6995139196515083, |
| "num_tokens": 26571893.0, |
| "step": 207 |
| }, |
| { |
| "entropy": 0.9895152673125267, |
| "epoch": 2.6688102893890675, |
| "grad_norm": 0.14651040732860565, |
| "learning_rate": 1.8027743175872664e-06, |
| "loss": 0.983437180519104, |
| "mean_token_accuracy": 0.7047524750232697, |
| "num_tokens": 26700860.0, |
| "step": 208 |
| }, |
| { |
| "entropy": 0.9806164056062698, |
| "epoch": 2.6816720257234725, |
| "grad_norm": 0.1419828236103058, |
| "learning_rate": 1.673191150434067e-06, |
| "loss": 0.9723012447357178, |
| "mean_token_accuracy": 0.7045335248112679, |
| "num_tokens": 26830268.0, |
| "step": 209 |
| }, |
| { |
| "entropy": 1.003640703856945, |
| "epoch": 2.694533762057878, |
| "grad_norm": 0.26877644658088684, |
| "learning_rate": 1.5482793275958529e-06, |
| "loss": 0.9935470819473267, |
| "mean_token_accuracy": 0.6985340863466263, |
| "num_tokens": 26959109.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.0105368196964264, |
| "epoch": 2.707395498392283, |
| "grad_norm": 0.16719378530979156, |
| "learning_rate": 1.428063863472895e-06, |
| "loss": 0.998131275177002, |
| "mean_token_accuracy": 0.6997055560350418, |
| "num_tokens": 27088875.0, |
| "step": 211 |
| }, |
| { |
| "entropy": 1.001611903309822, |
| "epoch": 2.720257234726688, |
| "grad_norm": 0.14638130366802216, |
| "learning_rate": 1.312568831989258e-06, |
| "loss": 0.9942350387573242, |
| "mean_token_accuracy": 0.6993389204144478, |
| "num_tokens": 27218005.0, |
| "step": 212 |
| }, |
| { |
| "entropy": 0.9718720093369484, |
| "epoch": 2.733118971061093, |
| "grad_norm": 0.1539345532655716, |
| "learning_rate": 1.201817361771837e-06, |
| "loss": 0.9664239287376404, |
| "mean_token_accuracy": 0.707412488758564, |
| "num_tokens": 27345118.0, |
| "step": 213 |
| }, |
| { |
| "entropy": 0.9780362546443939, |
| "epoch": 2.7459807073954985, |
| "grad_norm": 0.14015953242778778, |
| "learning_rate": 1.095831631518729e-06, |
| "loss": 0.9685409069061279, |
| "mean_token_accuracy": 0.706133060157299, |
| "num_tokens": 27473155.0, |
| "step": 214 |
| }, |
| { |
| "entropy": 0.9646907821297646, |
| "epoch": 2.7588424437299035, |
| "grad_norm": 0.14137625694274902, |
| "learning_rate": 9.946328655577624e-07, |
| "loss": 0.9556713700294495, |
| "mean_token_accuracy": 0.7085167616605759, |
| "num_tokens": 27602992.0, |
| "step": 215 |
| }, |
| { |
| "entropy": 0.969658762216568, |
| "epoch": 2.7717041800643085, |
| "grad_norm": 0.14186272025108337, |
| "learning_rate": 8.98241329596236e-07, |
| "loss": 0.963873565196991, |
| "mean_token_accuracy": 0.705694705247879, |
| "num_tokens": 27732425.0, |
| "step": 216 |
| }, |
| { |
| "entropy": 0.9947944805026054, |
| "epoch": 2.784565916398714, |
| "grad_norm": 0.14323608577251434, |
| "learning_rate": 8.066763266625282e-07, |
| "loss": 0.9879176020622253, |
| "mean_token_accuracy": 0.7011514753103256, |
| "num_tokens": 27861591.0, |
| "step": 217 |
| }, |
| { |
| "entropy": 0.9838699772953987, |
| "epoch": 2.797427652733119, |
| "grad_norm": 0.14354315400123596, |
| "learning_rate": 7.199561932405952e-07, |
| "loss": 0.9724022150039673, |
| "mean_token_accuracy": 0.7043187394738197, |
| "num_tokens": 27991145.0, |
| "step": 218 |
| }, |
| { |
| "entropy": 0.9758647233247757, |
| "epoch": 2.810289389067524, |
| "grad_norm": 0.14615927636623383, |
| "learning_rate": 6.380982955979192e-07, |
| "loss": 0.9710283279418945, |
| "mean_token_accuracy": 0.7058038115501404, |
| "num_tokens": 28121049.0, |
| "step": 219 |
| }, |
| { |
| "entropy": 0.9719097539782524, |
| "epoch": 2.823151125401929, |
| "grad_norm": 0.15186373889446259, |
| "learning_rate": 5.611190263078464e-07, |
| "loss": 0.9629349708557129, |
| "mean_token_accuracy": 0.7060469463467598, |
| "num_tokens": 28251006.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.9855683818459511, |
| "epoch": 2.8360128617363345, |
| "grad_norm": 0.1452024132013321, |
| "learning_rate": 4.890338009668316e-07, |
| "loss": 0.9732274413108826, |
| "mean_token_accuracy": 0.7041314393281937, |
| "num_tokens": 28380271.0, |
| "step": 221 |
| }, |
| { |
| "entropy": 0.9635618776082993, |
| "epoch": 2.8488745980707395, |
| "grad_norm": 0.14456729590892792, |
| "learning_rate": 4.2185705510739415e-07, |
| "loss": 0.9557806253433228, |
| "mean_token_accuracy": 0.7088164985179901, |
| "num_tokens": 28510489.0, |
| "step": 222 |
| }, |
| { |
| "entropy": 1.0085260719060898, |
| "epoch": 2.861736334405145, |
| "grad_norm": 0.1431344598531723, |
| "learning_rate": 3.5960224130728857e-07, |
| "loss": 1.0044658184051514, |
| "mean_token_accuracy": 0.696587473154068, |
| "num_tokens": 28638550.0, |
| "step": 223 |
| }, |
| { |
| "entropy": 1.0058645457029343, |
| "epoch": 2.87459807073955, |
| "grad_norm": 0.1409962773323059, |
| "learning_rate": 3.0228182649555126e-07, |
| "loss": 0.9938859939575195, |
| "mean_token_accuracy": 0.7006662487983704, |
| "num_tokens": 28768669.0, |
| "step": 224 |
| }, |
| { |
| "entropy": 0.9956894516944885, |
| "epoch": 2.887459807073955, |
| "grad_norm": 0.14797310531139374, |
| "learning_rate": 2.499072894559057e-07, |
| "loss": 0.9848740696907043, |
| "mean_token_accuracy": 0.7025537341833115, |
| "num_tokens": 28897003.0, |
| "step": 225 |
| }, |
| { |
| "entropy": 0.9903870970010757, |
| "epoch": 2.90032154340836, |
| "grad_norm": 0.14364922046661377, |
| "learning_rate": 2.0248911852807916e-07, |
| "loss": 0.9792647957801819, |
| "mean_token_accuracy": 0.7026231661438942, |
| "num_tokens": 29022883.0, |
| "step": 226 |
| }, |
| { |
| "entropy": 0.9798277169466019, |
| "epoch": 2.913183279742765, |
| "grad_norm": 0.14528293907642365, |
| "learning_rate": 1.6003680950742728e-07, |
| "loss": 0.9743708372116089, |
| "mean_token_accuracy": 0.7052473574876785, |
| "num_tokens": 29150724.0, |
| "step": 227 |
| }, |
| { |
| "entropy": 1.003961905837059, |
| "epoch": 2.9260450160771705, |
| "grad_norm": 0.14702223241329193, |
| "learning_rate": 1.2255886374334946e-07, |
| "loss": 1.0036938190460205, |
| "mean_token_accuracy": 0.6962004378437996, |
| "num_tokens": 29280359.0, |
| "step": 228 |
| }, |
| { |
| "entropy": 0.9913759529590607, |
| "epoch": 2.9389067524115755, |
| "grad_norm": 0.14192230999469757, |
| "learning_rate": 9.006278643683696e-08, |
| "loss": 0.9827680587768555, |
| "mean_token_accuracy": 0.7011790797114372, |
| "num_tokens": 29410482.0, |
| "step": 229 |
| }, |
| { |
| "entropy": 0.9629239290952682, |
| "epoch": 2.951768488745981, |
| "grad_norm": 0.14842632412910461, |
| "learning_rate": 6.255508513750841e-08, |
| "loss": 0.954397439956665, |
| "mean_token_accuracy": 0.708974651992321, |
| "num_tokens": 29540979.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.9984769746661186, |
| "epoch": 2.964630225080386, |
| "grad_norm": 0.14408619701862335, |
| "learning_rate": 4.004126844042444e-08, |
| "loss": 0.9900475740432739, |
| "mean_token_accuracy": 0.699671782553196, |
| "num_tokens": 29669764.0, |
| "step": 231 |
| }, |
| { |
| "entropy": 0.9829242154955864, |
| "epoch": 2.977491961414791, |
| "grad_norm": 0.1520869880914688, |
| "learning_rate": 2.2525844882964607e-08, |
| "loss": 0.9738559126853943, |
| "mean_token_accuracy": 0.7055443450808525, |
| "num_tokens": 29798608.0, |
| "step": 232 |
| }, |
| { |
| "entropy": 0.9934249892830849, |
| "epoch": 2.990353697749196, |
| "grad_norm": 0.14561671018600464, |
| "learning_rate": 1.0012322041960676e-08, |
| "loss": 0.983726978302002, |
| "mean_token_accuracy": 0.7004626914858818, |
| "num_tokens": 29927865.0, |
| "step": 233 |
| }, |
| { |
| "entropy": 0.9948156376679739, |
| "epoch": 3.0, |
| "grad_norm": 0.17473426461219788, |
| "learning_rate": 2.503205831277944e-09, |
| "loss": 0.9984904527664185, |
| "mean_token_accuracy": 0.6967108050982157, |
| "num_tokens": 29999036.0, |
| "step": 234 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 234, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "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": 3.1722382507691213e+18, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
