| { |
| "best_global_step": 3200, |
| "best_metric": 0.7580673098564148, |
| "best_model_checkpoint": "./gemma2b-indian-law-checkpoints/checkpoint-3200", |
| "epoch": 3.0, |
| "eval_steps": 200, |
| "global_step": 4908, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.012224938875305624, |
| "grad_norm": 37.56128692626953, |
| "learning_rate": 3.6e-05, |
| "loss": 10.4173, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.02444987775061125, |
| "grad_norm": 3.299354314804077, |
| "learning_rate": 7.6e-05, |
| "loss": 3.1786, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.03667481662591687, |
| "grad_norm": 0.9982728958129883, |
| "learning_rate": 0.000116, |
| "loss": 1.5718, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.0488997555012225, |
| "grad_norm": 1.614855170249939, |
| "learning_rate": 0.00015600000000000002, |
| "loss": 1.191, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.061124694376528114, |
| "grad_norm": 1.314520001411438, |
| "learning_rate": 0.000196, |
| "loss": 1.0527, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.07334963325183375, |
| "grad_norm": 0.7312303185462952, |
| "learning_rate": 0.00019999308358831873, |
| "loss": 1.2408, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.08557457212713937, |
| "grad_norm": 1.0890365839004517, |
| "learning_rate": 0.0001999691762330298, |
| "loss": 1.0016, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.097799511002445, |
| "grad_norm": 1.0758998394012451, |
| "learning_rate": 0.00019992819662814442, |
| "loss": 1.0788, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.1100244498777506, |
| "grad_norm": 0.6657894253730774, |
| "learning_rate": 0.00019987015177196007, |
| "loss": 1.0196, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.12224938875305623, |
| "grad_norm": 1.0422972440719604, |
| "learning_rate": 0.000199795051577095, |
| "loss": 0.9095, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.12224938875305623, |
| "eval_loss": 1.0752133131027222, |
| "eval_runtime": 76.2942, |
| "eval_samples_per_second": 19.071, |
| "eval_steps_per_second": 9.542, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.13447432762836187, |
| "grad_norm": 0.7305464148521423, |
| "learning_rate": 0.0001997029088687951, |
| "loss": 1.1615, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.1466992665036675, |
| "grad_norm": 0.6666431427001953, |
| "learning_rate": 0.00019959373938274383, |
| "loss": 0.931, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.15892420537897312, |
| "grad_norm": 0.7137753963470459, |
| "learning_rate": 0.00019946756176237503, |
| "loss": 1.0514, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.17114914425427874, |
| "grad_norm": 0.6486335396766663, |
| "learning_rate": 0.0001993243975556888, |
| "loss": 1.0359, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.18337408312958436, |
| "grad_norm": 0.7906181216239929, |
| "learning_rate": 0.0001991642712115719, |
| "loss": 0.8889, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.19559902200489, |
| "grad_norm": 0.6849639415740967, |
| "learning_rate": 0.00019898721007562248, |
| "loss": 1.1372, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.2078239608801956, |
| "grad_norm": 0.8779283165931702, |
| "learning_rate": 0.00019879324438547998, |
| "loss": 0.9307, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.2200488997555012, |
| "grad_norm": 0.7829409837722778, |
| "learning_rate": 0.00019858240726566132, |
| "loss": 0.9918, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.23227383863080683, |
| "grad_norm": 0.6620106101036072, |
| "learning_rate": 0.0001983547347219042, |
| "loss": 0.9512, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.24449877750611246, |
| "grad_norm": 0.7634586691856384, |
| "learning_rate": 0.0001981102656350181, |
| "loss": 0.8417, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.24449877750611246, |
| "eval_loss": 0.9910431504249573, |
| "eval_runtime": 73.7535, |
| "eval_samples_per_second": 19.728, |
| "eval_steps_per_second": 9.871, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.2567237163814181, |
| "grad_norm": 0.7715067267417908, |
| "learning_rate": 0.00019784904175424432, |
| "loss": 1.1179, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.26894865525672373, |
| "grad_norm": 0.672273576259613, |
| "learning_rate": 0.00019757110769012655, |
| "loss": 0.8962, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.28117359413202936, |
| "grad_norm": 0.7818410992622375, |
| "learning_rate": 0.00019727651090689215, |
| "loss": 0.9858, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.293398533007335, |
| "grad_norm": 0.7473219037055969, |
| "learning_rate": 0.0001969653017143466, |
| "loss": 0.9626, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.3056234718826406, |
| "grad_norm": 0.9187808036804199, |
| "learning_rate": 0.00019663753325928194, |
| "loss": 0.834, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.31784841075794623, |
| "grad_norm": 0.6319047808647156, |
| "learning_rate": 0.00019629326151640036, |
| "loss": 1.0821, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.33007334963325186, |
| "grad_norm": 0.8049977421760559, |
| "learning_rate": 0.00019593254527875534, |
| "loss": 0.8976, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.3422982885085575, |
| "grad_norm": 0.823184609413147, |
| "learning_rate": 0.0001955554461477112, |
| "loss": 0.9716, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.3545232273838631, |
| "grad_norm": 0.6882979869842529, |
| "learning_rate": 0.00019516202852242303, |
| "loss": 0.9341, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.36674816625916873, |
| "grad_norm": 0.6218671202659607, |
| "learning_rate": 0.000194752359588839, |
| "loss": 0.8244, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.36674816625916873, |
| "eval_loss": 0.959819495677948, |
| "eval_runtime": 72.7684, |
| "eval_samples_per_second": 19.995, |
| "eval_steps_per_second": 10.004, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.37897310513447435, |
| "grad_norm": 0.7724825143814087, |
| "learning_rate": 0.00019432650930822676, |
| "loss": 1.0712, |
| "step": 620 |
| }, |
| { |
| "epoch": 0.39119804400978, |
| "grad_norm": 0.6892656087875366, |
| "learning_rate": 0.00019388455040522548, |
| "loss": 0.9042, |
| "step": 640 |
| }, |
| { |
| "epoch": 0.4034229828850856, |
| "grad_norm": 0.796056866645813, |
| "learning_rate": 0.00019342655835542671, |
| "loss": 0.9478, |
| "step": 660 |
| }, |
| { |
| "epoch": 0.4156479217603912, |
| "grad_norm": 0.688993513584137, |
| "learning_rate": 0.00019295261137248465, |
| "loss": 0.9279, |
| "step": 680 |
| }, |
| { |
| "epoch": 0.4278728606356968, |
| "grad_norm": 0.741526186466217, |
| "learning_rate": 0.00019246279039475948, |
| "loss": 0.8097, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.4400977995110024, |
| "grad_norm": 0.7050778865814209, |
| "learning_rate": 0.00019195717907149492, |
| "loss": 1.0892, |
| "step": 720 |
| }, |
| { |
| "epoch": 0.45232273838630804, |
| "grad_norm": 0.5874431133270264, |
| "learning_rate": 0.0001914358637485331, |
| "loss": 0.8884, |
| "step": 740 |
| }, |
| { |
| "epoch": 0.46454767726161367, |
| "grad_norm": 0.8296729922294617, |
| "learning_rate": 0.0001908989334535688, |
| "loss": 0.9486, |
| "step": 760 |
| }, |
| { |
| "epoch": 0.4767726161369193, |
| "grad_norm": 0.7144893407821655, |
| "learning_rate": 0.00019034647988094586, |
| "loss": 0.9189, |
| "step": 780 |
| }, |
| { |
| "epoch": 0.4889975550122249, |
| "grad_norm": 0.772997260093689, |
| "learning_rate": 0.00018977859737599788, |
| "loss": 0.793, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.4889975550122249, |
| "eval_loss": 0.925308883190155, |
| "eval_runtime": 72.9139, |
| "eval_samples_per_second": 19.955, |
| "eval_steps_per_second": 9.984, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.5012224938875306, |
| "grad_norm": 0.7750535011291504, |
| "learning_rate": 0.00018919538291893643, |
| "loss": 1.0079, |
| "step": 820 |
| }, |
| { |
| "epoch": 0.5134474327628362, |
| "grad_norm": 0.9797669649124146, |
| "learning_rate": 0.00018859693610828935, |
| "loss": 0.8359, |
| "step": 840 |
| }, |
| { |
| "epoch": 0.5256723716381418, |
| "grad_norm": 0.8336035013198853, |
| "learning_rate": 0.00018798335914389167, |
| "loss": 0.9665, |
| "step": 860 |
| }, |
| { |
| "epoch": 0.5378973105134475, |
| "grad_norm": 0.7556520700454712, |
| "learning_rate": 0.00018735475680943251, |
| "loss": 0.9458, |
| "step": 880 |
| }, |
| { |
| "epoch": 0.5501222493887531, |
| "grad_norm": 0.599229097366333, |
| "learning_rate": 0.0001867112364545606, |
| "loss": 0.7951, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.5623471882640587, |
| "grad_norm": 0.599793016910553, |
| "learning_rate": 0.00018605290797655158, |
| "loss": 1.0371, |
| "step": 920 |
| }, |
| { |
| "epoch": 0.5745721271393643, |
| "grad_norm": 0.7806088924407959, |
| "learning_rate": 0.00018537988380154036, |
| "loss": 0.8426, |
| "step": 940 |
| }, |
| { |
| "epoch": 0.58679706601467, |
| "grad_norm": 0.7730845212936401, |
| "learning_rate": 0.00018469227886532149, |
| "loss": 0.9527, |
| "step": 960 |
| }, |
| { |
| "epoch": 0.5990220048899756, |
| "grad_norm": 0.698032796382904, |
| "learning_rate": 0.00018399021059372102, |
| "loss": 0.8948, |
| "step": 980 |
| }, |
| { |
| "epoch": 0.6112469437652812, |
| "grad_norm": 4.275924205780029, |
| "learning_rate": 0.00018327379888254293, |
| "loss": 0.8151, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.6112469437652812, |
| "eval_loss": 0.9073091149330139, |
| "eval_runtime": 73.0466, |
| "eval_samples_per_second": 19.919, |
| "eval_steps_per_second": 9.966, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.6234718826405868, |
| "grad_norm": 0.6735545992851257, |
| "learning_rate": 0.0001825431660770942, |
| "loss": 1.0144, |
| "step": 1020 |
| }, |
| { |
| "epoch": 0.6356968215158925, |
| "grad_norm": 0.7560144662857056, |
| "learning_rate": 0.00018179843695129098, |
| "loss": 0.8616, |
| "step": 1040 |
| }, |
| { |
| "epoch": 0.6479217603911981, |
| "grad_norm": 0.8442196846008301, |
| "learning_rate": 0.00018103973868635053, |
| "loss": 0.9166, |
| "step": 1060 |
| }, |
| { |
| "epoch": 0.6601466992665037, |
| "grad_norm": 0.7117698192596436, |
| "learning_rate": 0.00018026720084907168, |
| "loss": 0.8628, |
| "step": 1080 |
| }, |
| { |
| "epoch": 0.6723716381418093, |
| "grad_norm": 0.8544563055038452, |
| "learning_rate": 0.0001794809553697081, |
| "loss": 0.7934, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.684596577017115, |
| "grad_norm": 0.6967006921768188, |
| "learning_rate": 0.0001786811365194379, |
| "loss": 1.0055, |
| "step": 1120 |
| }, |
| { |
| "epoch": 0.6968215158924206, |
| "grad_norm": 0.6788530349731445, |
| "learning_rate": 0.00017786788088743345, |
| "loss": 0.8184, |
| "step": 1140 |
| }, |
| { |
| "epoch": 0.7090464547677262, |
| "grad_norm": 0.9361255168914795, |
| "learning_rate": 0.00017704132735753535, |
| "loss": 0.9082, |
| "step": 1160 |
| }, |
| { |
| "epoch": 0.7212713936430318, |
| "grad_norm": 0.7070406675338745, |
| "learning_rate": 0.00017620161708453447, |
| "loss": 0.8909, |
| "step": 1180 |
| }, |
| { |
| "epoch": 0.7334963325183375, |
| "grad_norm": 0.7315816283226013, |
| "learning_rate": 0.00017534889347006635, |
| "loss": 0.7943, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.7334963325183375, |
| "eval_loss": 0.8963550329208374, |
| "eval_runtime": 76.4414, |
| "eval_samples_per_second": 19.034, |
| "eval_steps_per_second": 9.524, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.7457212713936431, |
| "grad_norm": 0.6883059740066528, |
| "learning_rate": 0.00017448330213812153, |
| "loss": 0.9906, |
| "step": 1220 |
| }, |
| { |
| "epoch": 0.7579462102689487, |
| "grad_norm": 0.7072532773017883, |
| "learning_rate": 0.0001736049909101769, |
| "loss": 0.8023, |
| "step": 1240 |
| }, |
| { |
| "epoch": 0.7701711491442543, |
| "grad_norm": 0.8744392991065979, |
| "learning_rate": 0.00017271410977995122, |
| "loss": 0.9405, |
| "step": 1260 |
| }, |
| { |
| "epoch": 0.78239608801956, |
| "grad_norm": 0.8064141273498535, |
| "learning_rate": 0.00017181081088779002, |
| "loss": 0.8949, |
| "step": 1280 |
| }, |
| { |
| "epoch": 0.7946210268948656, |
| "grad_norm": 0.5958486199378967, |
| "learning_rate": 0.00017089524849468388, |
| "loss": 0.751, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.8068459657701712, |
| "grad_norm": 0.7758407592773438, |
| "learning_rate": 0.00016996757895592428, |
| "loss": 1.027, |
| "step": 1320 |
| }, |
| { |
| "epoch": 0.8190709046454768, |
| "grad_norm": 1.1253817081451416, |
| "learning_rate": 0.00016902796069440226, |
| "loss": 0.8156, |
| "step": 1340 |
| }, |
| { |
| "epoch": 0.8312958435207825, |
| "grad_norm": 0.7773277759552002, |
| "learning_rate": 0.0001680765541735535, |
| "loss": 0.8726, |
| "step": 1360 |
| }, |
| { |
| "epoch": 0.843520782396088, |
| "grad_norm": 0.843968391418457, |
| "learning_rate": 0.00016711352186995538, |
| "loss": 0.8573, |
| "step": 1380 |
| }, |
| { |
| "epoch": 0.8557457212713936, |
| "grad_norm": 0.5488795042037964, |
| "learning_rate": 0.00016613902824557983, |
| "loss": 0.7547, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.8557457212713936, |
| "eval_loss": 0.8634077906608582, |
| "eval_runtime": 73.8846, |
| "eval_samples_per_second": 19.693, |
| "eval_steps_per_second": 9.853, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.8679706601466992, |
| "grad_norm": 0.7018039226531982, |
| "learning_rate": 0.00016515323971970735, |
| "loss": 0.9663, |
| "step": 1420 |
| }, |
| { |
| "epoch": 0.8801955990220048, |
| "grad_norm": 0.8201090693473816, |
| "learning_rate": 0.0001641563246405069, |
| "loss": 0.8032, |
| "step": 1440 |
| }, |
| { |
| "epoch": 0.8924205378973105, |
| "grad_norm": 0.9070565104484558, |
| "learning_rate": 0.00016314845325628591, |
| "loss": 0.8828, |
| "step": 1460 |
| }, |
| { |
| "epoch": 0.9046454767726161, |
| "grad_norm": 0.8657905459403992, |
| "learning_rate": 0.00016212979768641644, |
| "loss": 0.8527, |
| "step": 1480 |
| }, |
| { |
| "epoch": 0.9168704156479217, |
| "grad_norm": 0.7145264744758606, |
| "learning_rate": 0.00016110053189194128, |
| "loss": 0.7485, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.9290953545232273, |
| "grad_norm": 0.8282883167266846, |
| "learning_rate": 0.00016006083164586565, |
| "loss": 0.9782, |
| "step": 1520 |
| }, |
| { |
| "epoch": 0.941320293398533, |
| "grad_norm": 0.6916366815567017, |
| "learning_rate": 0.0001590108745031398, |
| "loss": 0.8207, |
| "step": 1540 |
| }, |
| { |
| "epoch": 0.9535452322738386, |
| "grad_norm": 0.8130978345870972, |
| "learning_rate": 0.0001579508397703368, |
| "loss": 0.8883, |
| "step": 1560 |
| }, |
| { |
| "epoch": 0.9657701711491442, |
| "grad_norm": 0.6823825836181641, |
| "learning_rate": 0.00015688090847503164, |
| "loss": 0.8414, |
| "step": 1580 |
| }, |
| { |
| "epoch": 0.9779951100244498, |
| "grad_norm": 0.7315614223480225, |
| "learning_rate": 0.00015580126333488594, |
| "loss": 0.7448, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.9779951100244498, |
| "eval_loss": 0.8495904207229614, |
| "eval_runtime": 73.4994, |
| "eval_samples_per_second": 19.796, |
| "eval_steps_per_second": 9.905, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.9902200488997555, |
| "grad_norm": 0.8070491552352905, |
| "learning_rate": 0.0001547120887264446, |
| "loss": 0.9697, |
| "step": 1620 |
| }, |
| { |
| "epoch": 1.0024449877750612, |
| "grad_norm": 0.9631539583206177, |
| "learning_rate": 0.00015361357065364877, |
| "loss": 0.7887, |
| "step": 1640 |
| }, |
| { |
| "epoch": 1.0146699266503667, |
| "grad_norm": 0.8359513878822327, |
| "learning_rate": 0.00015250589671607094, |
| "loss": 0.804, |
| "step": 1660 |
| }, |
| { |
| "epoch": 1.0268948655256724, |
| "grad_norm": 0.9809061884880066, |
| "learning_rate": 0.00015138925607687762, |
| "loss": 0.6757, |
| "step": 1680 |
| }, |
| { |
| "epoch": 1.039119804400978, |
| "grad_norm": 0.7365483641624451, |
| "learning_rate": 0.0001502638394305251, |
| "loss": 0.7739, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.0513447432762837, |
| "grad_norm": 0.8231777548789978, |
| "learning_rate": 0.00014912983897019338, |
| "loss": 0.7197, |
| "step": 1720 |
| }, |
| { |
| "epoch": 1.0635696821515892, |
| "grad_norm": 1.193585991859436, |
| "learning_rate": 0.00014798744835496454, |
| "loss": 0.7066, |
| "step": 1740 |
| }, |
| { |
| "epoch": 1.075794621026895, |
| "grad_norm": 0.8090522289276123, |
| "learning_rate": 0.00014683686267675044, |
| "loss": 0.7726, |
| "step": 1760 |
| }, |
| { |
| "epoch": 1.0880195599022005, |
| "grad_norm": 0.9394134879112244, |
| "learning_rate": 0.00014567827842697592, |
| "loss": 0.6817, |
| "step": 1780 |
| }, |
| { |
| "epoch": 1.1002444987775062, |
| "grad_norm": 0.8854631185531616, |
| "learning_rate": 0.00014451189346302302, |
| "loss": 0.8068, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.1002444987775062, |
| "eval_loss": 0.8351919054985046, |
| "eval_runtime": 73.5928, |
| "eval_samples_per_second": 19.771, |
| "eval_steps_per_second": 9.892, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.1124694376528117, |
| "grad_norm": 0.7774815559387207, |
| "learning_rate": 0.00014333790697444176, |
| "loss": 0.6904, |
| "step": 1820 |
| }, |
| { |
| "epoch": 1.1246943765281174, |
| "grad_norm": 1.0627630949020386, |
| "learning_rate": 0.00014215651944893355, |
| "loss": 0.7048, |
| "step": 1840 |
| }, |
| { |
| "epoch": 1.136919315403423, |
| "grad_norm": 0.8762012124061584, |
| "learning_rate": 0.00014096793263811288, |
| "loss": 0.768, |
| "step": 1860 |
| }, |
| { |
| "epoch": 1.1491442542787287, |
| "grad_norm": 0.7565830945968628, |
| "learning_rate": 0.0001397723495230532, |
| "loss": 0.6863, |
| "step": 1880 |
| }, |
| { |
| "epoch": 1.1613691931540342, |
| "grad_norm": 0.9732739925384521, |
| "learning_rate": 0.00013856997427962273, |
| "loss": 0.768, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.17359413202934, |
| "grad_norm": 0.969767153263092, |
| "learning_rate": 0.00013736101224361635, |
| "loss": 0.7059, |
| "step": 1920 |
| }, |
| { |
| "epoch": 1.1858190709046454, |
| "grad_norm": 1.1064813137054443, |
| "learning_rate": 0.00013614566987568932, |
| "loss": 0.6862, |
| "step": 1940 |
| }, |
| { |
| "epoch": 1.1980440097799512, |
| "grad_norm": 0.9539498686790466, |
| "learning_rate": 0.00013492415472609896, |
| "loss": 0.7746, |
| "step": 1960 |
| }, |
| { |
| "epoch": 1.2102689486552567, |
| "grad_norm": 0.8654190897941589, |
| "learning_rate": 0.0001336966753992602, |
| "loss": 0.6644, |
| "step": 1980 |
| }, |
| { |
| "epoch": 1.2224938875305624, |
| "grad_norm": 0.7981263399124146, |
| "learning_rate": 0.00013246344151812114, |
| "loss": 0.7867, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.2224938875305624, |
| "eval_loss": 0.8208021521568298, |
| "eval_runtime": 73.6678, |
| "eval_samples_per_second": 19.751, |
| "eval_steps_per_second": 9.882, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.234718826405868, |
| "grad_norm": 0.855848491191864, |
| "learning_rate": 0.0001312246636883647, |
| "loss": 0.6893, |
| "step": 2020 |
| }, |
| { |
| "epoch": 1.2469437652811737, |
| "grad_norm": 1.0332973003387451, |
| "learning_rate": 0.00012998055346244235, |
| "loss": 0.6954, |
| "step": 2040 |
| }, |
| { |
| "epoch": 1.2591687041564792, |
| "grad_norm": 0.8407679796218872, |
| "learning_rate": 0.00012873132330344623, |
| "loss": 0.762, |
| "step": 2060 |
| }, |
| { |
| "epoch": 1.271393643031785, |
| "grad_norm": 1.0258105993270874, |
| "learning_rate": 0.00012747718654882576, |
| "loss": 0.6598, |
| "step": 2080 |
| }, |
| { |
| "epoch": 1.2836185819070904, |
| "grad_norm": 0.925604522228241, |
| "learning_rate": 0.0001262183573739549, |
| "loss": 0.7905, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.295843520782396, |
| "grad_norm": 0.8191604614257812, |
| "learning_rate": 0.0001249550507555563, |
| "loss": 0.6861, |
| "step": 2120 |
| }, |
| { |
| "epoch": 1.3080684596577017, |
| "grad_norm": 1.0716524124145508, |
| "learning_rate": 0.00012368748243498867, |
| "loss": 0.6863, |
| "step": 2140 |
| }, |
| { |
| "epoch": 1.3202933985330074, |
| "grad_norm": 0.9850893616676331, |
| "learning_rate": 0.0001224158688814034, |
| "loss": 0.7357, |
| "step": 2160 |
| }, |
| { |
| "epoch": 1.332518337408313, |
| "grad_norm": 0.9057893753051758, |
| "learning_rate": 0.00012114042725477714, |
| "loss": 0.6808, |
| "step": 2180 |
| }, |
| { |
| "epoch": 1.3447432762836184, |
| "grad_norm": 0.9190845489501953, |
| "learning_rate": 0.0001198613753688263, |
| "loss": 0.7383, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.3447432762836184, |
| "eval_loss": 0.8095206618309021, |
| "eval_runtime": 72.6766, |
| "eval_samples_per_second": 20.02, |
| "eval_steps_per_second": 10.017, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.3569682151589242, |
| "grad_norm": 0.8519930839538574, |
| "learning_rate": 0.00011857893165380967, |
| "loss": 0.7034, |
| "step": 2220 |
| }, |
| { |
| "epoch": 1.36919315403423, |
| "grad_norm": 1.1432406902313232, |
| "learning_rate": 0.00011729331511922623, |
| "loss": 0.6707, |
| "step": 2240 |
| }, |
| { |
| "epoch": 1.3814180929095354, |
| "grad_norm": 0.9781808257102966, |
| "learning_rate": 0.00011600474531641363, |
| "loss": 0.7653, |
| "step": 2260 |
| }, |
| { |
| "epoch": 1.393643031784841, |
| "grad_norm": 1.0139797925949097, |
| "learning_rate": 0.00011471344230105425, |
| "loss": 0.6318, |
| "step": 2280 |
| }, |
| { |
| "epoch": 1.4058679706601467, |
| "grad_norm": 0.9355171918869019, |
| "learning_rate": 0.00011341962659559527, |
| "loss": 0.7578, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.4180929095354524, |
| "grad_norm": 0.9339278936386108, |
| "learning_rate": 0.00011212351915158881, |
| "loss": 0.6647, |
| "step": 2320 |
| }, |
| { |
| "epoch": 1.430317848410758, |
| "grad_norm": 1.0720759630203247, |
| "learning_rate": 0.00011082534131195898, |
| "loss": 0.6953, |
| "step": 2340 |
| }, |
| { |
| "epoch": 1.4425427872860634, |
| "grad_norm": 0.9045745730400085, |
| "learning_rate": 0.00010952531477320203, |
| "loss": 0.7455, |
| "step": 2360 |
| }, |
| { |
| "epoch": 1.4547677261613692, |
| "grad_norm": 0.892172634601593, |
| "learning_rate": 0.00010822366154752592, |
| "loss": 0.6757, |
| "step": 2380 |
| }, |
| { |
| "epoch": 1.466992665036675, |
| "grad_norm": 0.8761904835700989, |
| "learning_rate": 0.00010692060392493644, |
| "loss": 0.7643, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.466992665036675, |
| "eval_loss": 0.8034286499023438, |
| "eval_runtime": 72.6861, |
| "eval_samples_per_second": 20.018, |
| "eval_steps_per_second": 10.016, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.4792176039119804, |
| "grad_norm": 0.7930101156234741, |
| "learning_rate": 0.00010561636443527533, |
| "loss": 0.6895, |
| "step": 2420 |
| }, |
| { |
| "epoch": 1.491442542787286, |
| "grad_norm": 1.095381259918213, |
| "learning_rate": 0.00010431116581021793, |
| "loss": 0.6797, |
| "step": 2440 |
| }, |
| { |
| "epoch": 1.5036674816625917, |
| "grad_norm": 1.0515379905700684, |
| "learning_rate": 0.00010300523094523605, |
| "loss": 0.7215, |
| "step": 2460 |
| }, |
| { |
| "epoch": 1.5158924205378974, |
| "grad_norm": 1.0114387273788452, |
| "learning_rate": 0.00010169878286153287, |
| "loss": 0.6612, |
| "step": 2480 |
| }, |
| { |
| "epoch": 1.528117359413203, |
| "grad_norm": 0.9440443515777588, |
| "learning_rate": 0.00010039204466795668, |
| "loss": 0.7216, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.5403422982885084, |
| "grad_norm": 0.9531809091567993, |
| "learning_rate": 9.908523952289928e-05, |
| "loss": 0.6749, |
| "step": 2520 |
| }, |
| { |
| "epoch": 1.5525672371638142, |
| "grad_norm": 1.3718974590301514, |
| "learning_rate": 9.777859059618612e-05, |
| "loss": 0.6885, |
| "step": 2540 |
| }, |
| { |
| "epoch": 1.56479217603912, |
| "grad_norm": 0.9546376466751099, |
| "learning_rate": 9.647232103096445e-05, |
| "loss": 0.7341, |
| "step": 2560 |
| }, |
| { |
| "epoch": 1.5770171149144254, |
| "grad_norm": 0.7880102396011353, |
| "learning_rate": 9.516665390559599e-05, |
| "loss": 0.6558, |
| "step": 2580 |
| }, |
| { |
| "epoch": 1.589242053789731, |
| "grad_norm": 0.9145961999893188, |
| "learning_rate": 9.386181219556073e-05, |
| "loss": 0.7466, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.589242053789731, |
| "eval_loss": 0.7883641123771667, |
| "eval_runtime": 72.6796, |
| "eval_samples_per_second": 20.019, |
| "eval_steps_per_second": 10.017, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.6014669926650367, |
| "grad_norm": 0.9434027075767517, |
| "learning_rate": 9.255801873537823e-05, |
| "loss": 0.6892, |
| "step": 2620 |
| }, |
| { |
| "epoch": 1.6136919315403424, |
| "grad_norm": 1.1631275415420532, |
| "learning_rate": 9.125549618055304e-05, |
| "loss": 0.6762, |
| "step": 2640 |
| }, |
| { |
| "epoch": 1.625916870415648, |
| "grad_norm": 0.9607406854629517, |
| "learning_rate": 8.995446696955073e-05, |
| "loss": 0.7475, |
| "step": 2660 |
| }, |
| { |
| "epoch": 1.6381418092909534, |
| "grad_norm": 0.7695438265800476, |
| "learning_rate": 8.865515328581079e-05, |
| "loss": 0.6396, |
| "step": 2680 |
| }, |
| { |
| "epoch": 1.6503667481662592, |
| "grad_norm": 0.9911854863166809, |
| "learning_rate": 8.735777701980336e-05, |
| "loss": 0.7409, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.662591687041565, |
| "grad_norm": 0.8723199367523193, |
| "learning_rate": 8.60625597311358e-05, |
| "loss": 0.6724, |
| "step": 2720 |
| }, |
| { |
| "epoch": 1.6748166259168704, |
| "grad_norm": 0.9864949584007263, |
| "learning_rate": 8.476972261071581e-05, |
| "loss": 0.6853, |
| "step": 2740 |
| }, |
| { |
| "epoch": 1.687041564792176, |
| "grad_norm": 0.9467067718505859, |
| "learning_rate": 8.347948644297756e-05, |
| "loss": 0.7384, |
| "step": 2760 |
| }, |
| { |
| "epoch": 1.6992665036674817, |
| "grad_norm": 0.913598895072937, |
| "learning_rate": 8.219207156817714e-05, |
| "loss": 0.6483, |
| "step": 2780 |
| }, |
| { |
| "epoch": 1.7114914425427874, |
| "grad_norm": 1.0271236896514893, |
| "learning_rate": 8.0907697844764e-05, |
| "loss": 0.739, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.7114914425427874, |
| "eval_loss": 0.7750269174575806, |
| "eval_runtime": 73.0417, |
| "eval_samples_per_second": 19.92, |
| "eval_steps_per_second": 9.967, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.723716381418093, |
| "grad_norm": 1.0170419216156006, |
| "learning_rate": 7.962658461183446e-05, |
| "loss": 0.6752, |
| "step": 2820 |
| }, |
| { |
| "epoch": 1.7359413202933984, |
| "grad_norm": 1.1174966096878052, |
| "learning_rate": 7.834895065167429e-05, |
| "loss": 0.6594, |
| "step": 2840 |
| }, |
| { |
| "epoch": 1.7481662591687042, |
| "grad_norm": 0.7971872687339783, |
| "learning_rate": 7.707501415239584e-05, |
| "loss": 0.7223, |
| "step": 2860 |
| }, |
| { |
| "epoch": 1.76039119804401, |
| "grad_norm": 0.7204257845878601, |
| "learning_rate": 7.580499267067714e-05, |
| "loss": 0.6332, |
| "step": 2880 |
| }, |
| { |
| "epoch": 1.7726161369193154, |
| "grad_norm": 0.8263009786605835, |
| "learning_rate": 7.453910309460857e-05, |
| "loss": 0.7105, |
| "step": 2900 |
| }, |
| { |
| "epoch": 1.784841075794621, |
| "grad_norm": 0.8076934218406677, |
| "learning_rate": 7.327756160665377e-05, |
| "loss": 0.6418, |
| "step": 2920 |
| }, |
| { |
| "epoch": 1.7970660146699267, |
| "grad_norm": 1.0033643245697021, |
| "learning_rate": 7.202058364673097e-05, |
| "loss": 0.6737, |
| "step": 2940 |
| }, |
| { |
| "epoch": 1.8092909535452324, |
| "grad_norm": 0.9179111123085022, |
| "learning_rate": 7.076838387542136e-05, |
| "loss": 0.7182, |
| "step": 2960 |
| }, |
| { |
| "epoch": 1.821515892420538, |
| "grad_norm": 0.7466602921485901, |
| "learning_rate": 6.95211761373103e-05, |
| "loss": 0.6172, |
| "step": 2980 |
| }, |
| { |
| "epoch": 1.8337408312958434, |
| "grad_norm": 1.0629620552062988, |
| "learning_rate": 6.827917342446802e-05, |
| "loss": 0.7433, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.8337408312958434, |
| "eval_loss": 0.765312910079956, |
| "eval_runtime": 73.3067, |
| "eval_samples_per_second": 19.848, |
| "eval_steps_per_second": 9.931, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.8459657701711492, |
| "grad_norm": 0.8783513307571411, |
| "learning_rate": 6.704258784007584e-05, |
| "loss": 0.6621, |
| "step": 3020 |
| }, |
| { |
| "epoch": 1.858190709046455, |
| "grad_norm": 1.1947550773620605, |
| "learning_rate": 6.581163056220423e-05, |
| "loss": 0.645, |
| "step": 3040 |
| }, |
| { |
| "epoch": 1.8704156479217604, |
| "grad_norm": 0.9742974042892456, |
| "learning_rate": 6.458651180774886e-05, |
| "loss": 0.6833, |
| "step": 3060 |
| }, |
| { |
| "epoch": 1.882640586797066, |
| "grad_norm": 0.7691249847412109, |
| "learning_rate": 6.336744079653079e-05, |
| "loss": 0.6281, |
| "step": 3080 |
| }, |
| { |
| "epoch": 1.8948655256723717, |
| "grad_norm": 0.9411635398864746, |
| "learning_rate": 6.215462571556687e-05, |
| "loss": 0.7355, |
| "step": 3100 |
| }, |
| { |
| "epoch": 1.9070904645476774, |
| "grad_norm": 1.1614731550216675, |
| "learning_rate": 6.094827368351665e-05, |
| "loss": 0.6707, |
| "step": 3120 |
| }, |
| { |
| "epoch": 1.919315403422983, |
| "grad_norm": 1.0442848205566406, |
| "learning_rate": 5.9748590715311645e-05, |
| "loss": 0.6622, |
| "step": 3140 |
| }, |
| { |
| "epoch": 1.9315403422982884, |
| "grad_norm": 0.9230803847312927, |
| "learning_rate": 5.855578168697302e-05, |
| "loss": 0.7039, |
| "step": 3160 |
| }, |
| { |
| "epoch": 1.9437652811735942, |
| "grad_norm": 0.849789559841156, |
| "learning_rate": 5.7370050300623944e-05, |
| "loss": 0.6248, |
| "step": 3180 |
| }, |
| { |
| "epoch": 1.9559902200488999, |
| "grad_norm": 1.0967777967453003, |
| "learning_rate": 5.6191599049702426e-05, |
| "loss": 0.6955, |
| "step": 3200 |
| }, |
| { |
| "epoch": 1.9559902200488999, |
| "eval_loss": 0.7580673098564148, |
| "eval_runtime": 72.9299, |
| "eval_samples_per_second": 19.951, |
| "eval_steps_per_second": 9.982, |
| "step": 3200 |
| }, |
| { |
| "epoch": 1.9682151589242054, |
| "grad_norm": 0.800773561000824, |
| "learning_rate": 5.502062918438023e-05, |
| "loss": 0.6454, |
| "step": 3220 |
| }, |
| { |
| "epoch": 1.980440097799511, |
| "grad_norm": 1.2560752630233765, |
| "learning_rate": 5.385734067719457e-05, |
| "loss": 0.6369, |
| "step": 3240 |
| }, |
| { |
| "epoch": 1.9926650366748166, |
| "grad_norm": 0.9020423293113708, |
| "learning_rate": 5.270193218889758e-05, |
| "loss": 0.6938, |
| "step": 3260 |
| }, |
| { |
| "epoch": 2.0048899755501224, |
| "grad_norm": 0.8488172888755798, |
| "learning_rate": 5.1554601034530225e-05, |
| "loss": 0.5916, |
| "step": 3280 |
| }, |
| { |
| "epoch": 2.0171149144254277, |
| "grad_norm": 0.9031133651733398, |
| "learning_rate": 5.0415543149725585e-05, |
| "loss": 0.5356, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.0293398533007334, |
| "grad_norm": 0.8954846262931824, |
| "learning_rate": 4.9284953057248075e-05, |
| "loss": 0.5228, |
| "step": 3320 |
| }, |
| { |
| "epoch": 2.041564792176039, |
| "grad_norm": 0.9425753951072693, |
| "learning_rate": 4.816302383377368e-05, |
| "loss": 0.5642, |
| "step": 3340 |
| }, |
| { |
| "epoch": 2.053789731051345, |
| "grad_norm": 0.899931013584137, |
| "learning_rate": 4.704994707691735e-05, |
| "loss": 0.494, |
| "step": 3360 |
| }, |
| { |
| "epoch": 2.06601466992665, |
| "grad_norm": 1.266194224357605, |
| "learning_rate": 4.5945912872512806e-05, |
| "loss": 0.5377, |
| "step": 3380 |
| }, |
| { |
| "epoch": 2.078239608801956, |
| "grad_norm": 0.9894401431083679, |
| "learning_rate": 4.485110976215069e-05, |
| "loss": 0.5248, |
| "step": 3400 |
| }, |
| { |
| "epoch": 2.078239608801956, |
| "eval_loss": 0.782360315322876, |
| "eval_runtime": 73.2082, |
| "eval_samples_per_second": 19.875, |
| "eval_steps_per_second": 9.944, |
| "step": 3400 |
| }, |
| { |
| "epoch": 2.0904645476772616, |
| "grad_norm": 1.1742653846740723, |
| "learning_rate": 4.376572471098026e-05, |
| "loss": 0.515, |
| "step": 3420 |
| }, |
| { |
| "epoch": 2.1026894865525674, |
| "grad_norm": 0.8864428400993347, |
| "learning_rate": 4.2689943075780545e-05, |
| "loss": 0.5572, |
| "step": 3440 |
| }, |
| { |
| "epoch": 2.1149144254278727, |
| "grad_norm": 0.950120210647583, |
| "learning_rate": 4.162394857330582e-05, |
| "loss": 0.5034, |
| "step": 3460 |
| }, |
| { |
| "epoch": 2.1271393643031784, |
| "grad_norm": 1.3487776517868042, |
| "learning_rate": 4.0567923248911444e-05, |
| "loss": 0.564, |
| "step": 3480 |
| }, |
| { |
| "epoch": 2.139364303178484, |
| "grad_norm": 0.9111340045928955, |
| "learning_rate": 3.952204744546508e-05, |
| "loss": 0.5207, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.15158924205379, |
| "grad_norm": 1.182996392250061, |
| "learning_rate": 3.8486499772548565e-05, |
| "loss": 0.513, |
| "step": 3520 |
| }, |
| { |
| "epoch": 2.163814180929095, |
| "grad_norm": 1.0959248542785645, |
| "learning_rate": 3.746145707595584e-05, |
| "loss": 0.5577, |
| "step": 3540 |
| }, |
| { |
| "epoch": 2.176039119804401, |
| "grad_norm": 1.0401321649551392, |
| "learning_rate": 3.64470944074921e-05, |
| "loss": 0.5178, |
| "step": 3560 |
| }, |
| { |
| "epoch": 2.1882640586797066, |
| "grad_norm": 1.367658019065857, |
| "learning_rate": 3.544358499507933e-05, |
| "loss": 0.5655, |
| "step": 3580 |
| }, |
| { |
| "epoch": 2.2004889975550124, |
| "grad_norm": 0.9502832293510437, |
| "learning_rate": 3.4451100213173113e-05, |
| "loss": 0.5193, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.2004889975550124, |
| "eval_loss": 0.7838400602340698, |
| "eval_runtime": 73.3883, |
| "eval_samples_per_second": 19.826, |
| "eval_steps_per_second": 9.92, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.2127139364303177, |
| "grad_norm": 0.9085776805877686, |
| "learning_rate": 3.3469809553496424e-05, |
| "loss": 0.5115, |
| "step": 3620 |
| }, |
| { |
| "epoch": 2.2249388753056234, |
| "grad_norm": 1.031757116317749, |
| "learning_rate": 3.249988059609441e-05, |
| "loss": 0.5625, |
| "step": 3640 |
| }, |
| { |
| "epoch": 2.237163814180929, |
| "grad_norm": 1.1373658180236816, |
| "learning_rate": 3.154147898071601e-05, |
| "loss": 0.5095, |
| "step": 3660 |
| }, |
| { |
| "epoch": 2.249388753056235, |
| "grad_norm": 1.1333980560302734, |
| "learning_rate": 3.059476837852675e-05, |
| "loss": 0.5526, |
| "step": 3680 |
| }, |
| { |
| "epoch": 2.26161369193154, |
| "grad_norm": 0.9720286130905151, |
| "learning_rate": 2.9659910464157803e-05, |
| "loss": 0.5272, |
| "step": 3700 |
| }, |
| { |
| "epoch": 2.273838630806846, |
| "grad_norm": 0.9294410347938538, |
| "learning_rate": 2.8737064888095978e-05, |
| "loss": 0.5074, |
| "step": 3720 |
| }, |
| { |
| "epoch": 2.2860635696821516, |
| "grad_norm": 1.2121460437774658, |
| "learning_rate": 2.7826389249419383e-05, |
| "loss": 0.5449, |
| "step": 3740 |
| }, |
| { |
| "epoch": 2.2982885085574574, |
| "grad_norm": 0.9010898470878601, |
| "learning_rate": 2.6928039068883417e-05, |
| "loss": 0.5022, |
| "step": 3760 |
| }, |
| { |
| "epoch": 2.3105134474327627, |
| "grad_norm": 1.1063168048858643, |
| "learning_rate": 2.6042167762361714e-05, |
| "loss": 0.5523, |
| "step": 3780 |
| }, |
| { |
| "epoch": 2.3227383863080684, |
| "grad_norm": 1.0388875007629395, |
| "learning_rate": 2.5168926614646494e-05, |
| "loss": 0.5233, |
| "step": 3800 |
| }, |
| { |
| "epoch": 2.3227383863080684, |
| "eval_loss": 0.7787972092628479, |
| "eval_runtime": 72.9967, |
| "eval_samples_per_second": 19.932, |
| "eval_steps_per_second": 9.973, |
| "step": 3800 |
| }, |
| { |
| "epoch": 2.334963325183374, |
| "grad_norm": 0.8659378290176392, |
| "learning_rate": 2.4308464753612902e-05, |
| "loss": 0.4965, |
| "step": 3820 |
| }, |
| { |
| "epoch": 2.34718826405868, |
| "grad_norm": 1.1549756526947021, |
| "learning_rate": 2.3460929124751684e-05, |
| "loss": 0.5427, |
| "step": 3840 |
| }, |
| { |
| "epoch": 2.359413202933985, |
| "grad_norm": 0.8729878664016724, |
| "learning_rate": 2.2626464466074505e-05, |
| "loss": 0.4988, |
| "step": 3860 |
| }, |
| { |
| "epoch": 2.371638141809291, |
| "grad_norm": 1.442050814628601, |
| "learning_rate": 2.180521328339635e-05, |
| "loss": 0.5419, |
| "step": 3880 |
| }, |
| { |
| "epoch": 2.3838630806845966, |
| "grad_norm": 0.9110784530639648, |
| "learning_rate": 2.0997315825999077e-05, |
| "loss": 0.5111, |
| "step": 3900 |
| }, |
| { |
| "epoch": 2.3960880195599024, |
| "grad_norm": 1.0884591341018677, |
| "learning_rate": 2.020291006268026e-05, |
| "loss": 0.5076, |
| "step": 3920 |
| }, |
| { |
| "epoch": 2.4083129584352077, |
| "grad_norm": 1.425678014755249, |
| "learning_rate": 1.94221316581916e-05, |
| "loss": 0.5669, |
| "step": 3940 |
| }, |
| { |
| "epoch": 2.4205378973105134, |
| "grad_norm": 0.8123071789741516, |
| "learning_rate": 1.8655113950070767e-05, |
| "loss": 0.5086, |
| "step": 3960 |
| }, |
| { |
| "epoch": 2.432762836185819, |
| "grad_norm": 1.253361463546753, |
| "learning_rate": 1.7901987925870644e-05, |
| "loss": 0.533, |
| "step": 3980 |
| }, |
| { |
| "epoch": 2.444987775061125, |
| "grad_norm": 0.9195329546928406, |
| "learning_rate": 1.7162882200789908e-05, |
| "loss": 0.5122, |
| "step": 4000 |
| }, |
| { |
| "epoch": 2.444987775061125, |
| "eval_loss": 0.7787081003189087, |
| "eval_runtime": 73.3677, |
| "eval_samples_per_second": 19.832, |
| "eval_steps_per_second": 9.923, |
| "step": 4000 |
| }, |
| { |
| "epoch": 2.45721271393643, |
| "grad_norm": 1.03239107131958, |
| "learning_rate": 1.6437922995708755e-05, |
| "loss": 0.4915, |
| "step": 4020 |
| }, |
| { |
| "epoch": 2.469437652811736, |
| "grad_norm": 1.1267759799957275, |
| "learning_rate": 1.5727234115633506e-05, |
| "loss": 0.5327, |
| "step": 4040 |
| }, |
| { |
| "epoch": 2.4816625916870416, |
| "grad_norm": 0.9993512630462646, |
| "learning_rate": 1.5030936928553818e-05, |
| "loss": 0.4893, |
| "step": 4060 |
| }, |
| { |
| "epoch": 2.4938875305623474, |
| "grad_norm": 1.3921096324920654, |
| "learning_rate": 1.434915034471601e-05, |
| "loss": 0.5478, |
| "step": 4080 |
| }, |
| { |
| "epoch": 2.5061124694376526, |
| "grad_norm": 1.3431448936462402, |
| "learning_rate": 1.3681990796316202e-05, |
| "loss": 0.5224, |
| "step": 4100 |
| }, |
| { |
| "epoch": 2.5183374083129584, |
| "grad_norm": 0.9568518400192261, |
| "learning_rate": 1.3029572217616503e-05, |
| "loss": 0.5121, |
| "step": 4120 |
| }, |
| { |
| "epoch": 2.530562347188264, |
| "grad_norm": 0.952880322933197, |
| "learning_rate": 1.239200602548799e-05, |
| "loss": 0.5444, |
| "step": 4140 |
| }, |
| { |
| "epoch": 2.54278728606357, |
| "grad_norm": 0.9126360416412354, |
| "learning_rate": 1.176940110038336e-05, |
| "loss": 0.5003, |
| "step": 4160 |
| }, |
| { |
| "epoch": 2.555012224938875, |
| "grad_norm": 1.1036409139633179, |
| "learning_rate": 1.1161863767742941e-05, |
| "loss": 0.5475, |
| "step": 4180 |
| }, |
| { |
| "epoch": 2.567237163814181, |
| "grad_norm": 0.9962987899780273, |
| "learning_rate": 1.0569497779836912e-05, |
| "loss": 0.4948, |
| "step": 4200 |
| }, |
| { |
| "epoch": 2.567237163814181, |
| "eval_loss": 0.7748263478279114, |
| "eval_runtime": 73.9445, |
| "eval_samples_per_second": 19.677, |
| "eval_steps_per_second": 9.845, |
| "step": 4200 |
| }, |
| { |
| "epoch": 2.5794621026894866, |
| "grad_norm": 0.817188560962677, |
| "learning_rate": 9.992404298047008e-06, |
| "loss": 0.5096, |
| "step": 4220 |
| }, |
| { |
| "epoch": 2.591687041564792, |
| "grad_norm": 1.1667689085006714, |
| "learning_rate": 9.430681875590652e-06, |
| "loss": 0.5517, |
| "step": 4240 |
| }, |
| { |
| "epoch": 2.6039119804400976, |
| "grad_norm": 1.010845422744751, |
| "learning_rate": 8.88442644069053e-06, |
| "loss": 0.5099, |
| "step": 4260 |
| }, |
| { |
| "epoch": 2.6161369193154034, |
| "grad_norm": 1.498199701309204, |
| "learning_rate": 8.353731280192455e-06, |
| "loss": 0.5326, |
| "step": 4280 |
| }, |
| { |
| "epoch": 2.628361858190709, |
| "grad_norm": 1.0481910705566406, |
| "learning_rate": 7.838687023634262e-06, |
| "loss": 0.5049, |
| "step": 4300 |
| }, |
| { |
| "epoch": 2.640586797066015, |
| "grad_norm": 1.1396204233169556, |
| "learning_rate": 7.3393816277685375e-06, |
| "loss": 0.5049, |
| "step": 4320 |
| }, |
| { |
| "epoch": 2.65281173594132, |
| "grad_norm": 1.1356322765350342, |
| "learning_rate": 6.855900361541811e-06, |
| "loss": 0.5478, |
| "step": 4340 |
| }, |
| { |
| "epoch": 2.665036674816626, |
| "grad_norm": 0.8558567762374878, |
| "learning_rate": 6.388325791532768e-06, |
| "loss": 0.5117, |
| "step": 4360 |
| }, |
| { |
| "epoch": 2.6772616136919316, |
| "grad_norm": 1.3358498811721802, |
| "learning_rate": 5.936737767851818e-06, |
| "loss": 0.531, |
| "step": 4380 |
| }, |
| { |
| "epoch": 2.689486552567237, |
| "grad_norm": 0.9907200336456299, |
| "learning_rate": 5.501213410504824e-06, |
| "loss": 0.5135, |
| "step": 4400 |
| }, |
| { |
| "epoch": 2.689486552567237, |
| "eval_loss": 0.7737353444099426, |
| "eval_runtime": 73.516, |
| "eval_samples_per_second": 19.792, |
| "eval_steps_per_second": 9.903, |
| "step": 4400 |
| }, |
| { |
| "epoch": 2.7017114914425426, |
| "grad_norm": 0.8333169221878052, |
| "learning_rate": 5.081827096222824e-06, |
| "loss": 0.4998, |
| "step": 4420 |
| }, |
| { |
| "epoch": 2.7139364303178484, |
| "grad_norm": 1.1503171920776367, |
| "learning_rate": 4.678650445760391e-06, |
| "loss": 0.5693, |
| "step": 4440 |
| }, |
| { |
| "epoch": 2.726161369193154, |
| "grad_norm": 1.0463289022445679, |
| "learning_rate": 4.291752311664566e-06, |
| "loss": 0.5148, |
| "step": 4460 |
| }, |
| { |
| "epoch": 2.73838630806846, |
| "grad_norm": 1.0845234394073486, |
| "learning_rate": 3.9211987665165205e-06, |
| "loss": 0.5546, |
| "step": 4480 |
| }, |
| { |
| "epoch": 2.750611246943765, |
| "grad_norm": 1.271662712097168, |
| "learning_rate": 3.567053091648076e-06, |
| "loss": 0.5228, |
| "step": 4500 |
| }, |
| { |
| "epoch": 2.762836185819071, |
| "grad_norm": 0.9263620972633362, |
| "learning_rate": 3.2293757663347634e-06, |
| "loss": 0.4931, |
| "step": 4520 |
| }, |
| { |
| "epoch": 2.7750611246943766, |
| "grad_norm": 1.2969563007354736, |
| "learning_rate": 2.9082244574675123e-06, |
| "loss": 0.5519, |
| "step": 4540 |
| }, |
| { |
| "epoch": 2.787286063569682, |
| "grad_norm": 0.9190470576286316, |
| "learning_rate": 2.603654009704537e-06, |
| "loss": 0.4886, |
| "step": 4560 |
| }, |
| { |
| "epoch": 2.7995110024449876, |
| "grad_norm": 1.1626344919204712, |
| "learning_rate": 2.315716436105253e-06, |
| "loss": 0.535, |
| "step": 4580 |
| }, |
| { |
| "epoch": 2.8117359413202934, |
| "grad_norm": 1.2037278413772583, |
| "learning_rate": 2.0444609092477585e-06, |
| "loss": 0.507, |
| "step": 4600 |
| }, |
| { |
| "epoch": 2.8117359413202934, |
| "eval_loss": 0.7746286988258362, |
| "eval_runtime": 73.6079, |
| "eval_samples_per_second": 19.767, |
| "eval_steps_per_second": 9.89, |
| "step": 4600 |
| }, |
| { |
| "epoch": 2.823960880195599, |
| "grad_norm": 0.8642593026161194, |
| "learning_rate": 1.7899337528313409e-06, |
| "loss": 0.498, |
| "step": 4620 |
| }, |
| { |
| "epoch": 2.836185819070905, |
| "grad_norm": 1.236799716949463, |
| "learning_rate": 1.5521784337655943e-06, |
| "loss": 0.5443, |
| "step": 4640 |
| }, |
| { |
| "epoch": 2.84841075794621, |
| "grad_norm": 1.100149154663086, |
| "learning_rate": 1.3312355547472898e-06, |
| "loss": 0.4933, |
| "step": 4660 |
| }, |
| { |
| "epoch": 2.860635696821516, |
| "grad_norm": 1.3998684883117676, |
| "learning_rate": 1.1271428473264678e-06, |
| "loss": 0.5374, |
| "step": 4680 |
| }, |
| { |
| "epoch": 2.8728606356968216, |
| "grad_norm": 1.1776994466781616, |
| "learning_rate": 9.399351654628797e-07, |
| "loss": 0.5098, |
| "step": 4700 |
| }, |
| { |
| "epoch": 2.885085574572127, |
| "grad_norm": 0.8102136850357056, |
| "learning_rate": 7.696444795737057e-07, |
| "loss": 0.5033, |
| "step": 4720 |
| }, |
| { |
| "epoch": 2.8973105134474326, |
| "grad_norm": 1.111661434173584, |
| "learning_rate": 6.162998710738666e-07, |
| "loss": 0.5397, |
| "step": 4740 |
| }, |
| { |
| "epoch": 2.9095354523227384, |
| "grad_norm": 0.8676882982254028, |
| "learning_rate": 4.799275274096293e-07, |
| "loss": 0.5131, |
| "step": 4760 |
| }, |
| { |
| "epoch": 2.921760391198044, |
| "grad_norm": 1.209560513496399, |
| "learning_rate": 3.605507375864292e-07, |
| "loss": 0.533, |
| "step": 4780 |
| }, |
| { |
| "epoch": 2.93398533007335, |
| "grad_norm": 0.9141420722007751, |
| "learning_rate": 2.5818988819170667e-07, |
| "loss": 0.5137, |
| "step": 4800 |
| }, |
| { |
| "epoch": 2.93398533007335, |
| "eval_loss": 0.7741885185241699, |
| "eval_runtime": 73.2546, |
| "eval_samples_per_second": 19.862, |
| "eval_steps_per_second": 9.938, |
| "step": 4800 |
| }, |
| { |
| "epoch": 2.946210268948655, |
| "grad_norm": 1.0431549549102783, |
| "learning_rate": 1.7286245991336992e-07, |
| "loss": 0.4988, |
| "step": 4820 |
| }, |
| { |
| "epoch": 2.958435207823961, |
| "grad_norm": 1.0762169361114502, |
| "learning_rate": 1.0458302455450542e-07, |
| "loss": 0.5352, |
| "step": 4840 |
| }, |
| { |
| "epoch": 2.9706601466992666, |
| "grad_norm": 0.9684817790985107, |
| "learning_rate": 5.336324254495706e-08, |
| "loss": 0.5019, |
| "step": 4860 |
| }, |
| { |
| "epoch": 2.982885085574572, |
| "grad_norm": 1.2538353204727173, |
| "learning_rate": 1.921186094989702e-08, |
| "loss": 0.5196, |
| "step": 4880 |
| }, |
| { |
| "epoch": 2.9951100244498776, |
| "grad_norm": 1.1605544090270996, |
| "learning_rate": 2.1347119761649935e-09, |
| "loss": 0.4986, |
| "step": 4900 |
| } |
| ], |
| "logging_steps": 20, |
| "max_steps": 4908, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 200, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.896129323819827e+16, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
