| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 10.0, |
| "eval_steps": 500, |
| "global_step": 1890, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 2.961631564795971, |
| "epoch": 0.05291005291005291, |
| "grad_norm": 37.0, |
| "learning_rate": 3.2142857142857147e-06, |
| "loss": 3.414657974243164, |
| "mean_token_accuracy": 0.4348594455048442, |
| "num_tokens": 710093.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 2.68032029569149, |
| "epoch": 0.10582010582010581, |
| "grad_norm": 21.875, |
| "learning_rate": 6.785714285714287e-06, |
| "loss": 2.987566375732422, |
| "mean_token_accuracy": 0.4721761476248503, |
| "num_tokens": 1429271.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.8399402394890785, |
| "epoch": 0.15873015873015872, |
| "grad_norm": 6.90625, |
| "learning_rate": 1.0357142857142859e-05, |
| "loss": 1.9313663482666015, |
| "mean_token_accuracy": 0.5914886150509119, |
| "num_tokens": 2149735.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.315066960453987, |
| "epoch": 0.21164021164021163, |
| "grad_norm": 2.65625, |
| "learning_rate": 1.3928571428571429e-05, |
| "loss": 1.364548397064209, |
| "mean_token_accuracy": 0.6834412924945354, |
| "num_tokens": 2868867.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.0669520691037178, |
| "epoch": 0.26455026455026454, |
| "grad_norm": 1.984375, |
| "learning_rate": 1.7500000000000002e-05, |
| "loss": 1.0865763664245605, |
| "mean_token_accuracy": 0.7323449164628982, |
| "num_tokens": 3582476.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.9911199904978275, |
| "epoch": 0.31746031746031744, |
| "grad_norm": 2.0625, |
| "learning_rate": 1.999986795783372e-05, |
| "loss": 1.0018372535705566, |
| "mean_token_accuracy": 0.7486573837697506, |
| "num_tokens": 4287689.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.9219351679086685, |
| "epoch": 0.37037037037037035, |
| "grad_norm": 1.703125, |
| "learning_rate": 1.9997520638548385e-05, |
| "loss": 0.9412899017333984, |
| "mean_token_accuracy": 0.7611449956893921, |
| "num_tokens": 5011191.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.8958804056048393, |
| "epoch": 0.42328042328042326, |
| "grad_norm": 1.5703125, |
| "learning_rate": 1.9992239841688418e-05, |
| "loss": 0.903989028930664, |
| "mean_token_accuracy": 0.7697657376527787, |
| "num_tokens": 5714842.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.8621860913932323, |
| "epoch": 0.47619047619047616, |
| "grad_norm": 1.6484375, |
| "learning_rate": 1.9984027116747914e-05, |
| "loss": 0.8682960510253906, |
| "mean_token_accuracy": 0.7767987586557865, |
| "num_tokens": 6424614.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.8519016958773136, |
| "epoch": 0.5291005291005291, |
| "grad_norm": 1.6484375, |
| "learning_rate": 1.997288487350878e-05, |
| "loss": 0.8630915641784668, |
| "mean_token_accuracy": 0.7774669796228408, |
| "num_tokens": 7139778.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.8354873709380627, |
| "epoch": 0.582010582010582, |
| "grad_norm": 1.546875, |
| "learning_rate": 1.9958816381333684e-05, |
| "loss": 0.8412243843078613, |
| "mean_token_accuracy": 0.7815381772816181, |
| "num_tokens": 7852540.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.816736352443695, |
| "epoch": 0.6349206349206349, |
| "grad_norm": 1.4609375, |
| "learning_rate": 1.994182576820673e-05, |
| "loss": 0.8229312896728516, |
| "mean_token_accuracy": 0.7848087482154369, |
| "num_tokens": 8568056.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.8098116122186184, |
| "epoch": 0.6878306878306878, |
| "grad_norm": 1.6953125, |
| "learning_rate": 1.9921918019522238e-05, |
| "loss": 0.8195478439331054, |
| "mean_token_accuracy": 0.786376665532589, |
| "num_tokens": 9276742.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.7951123341917992, |
| "epoch": 0.7407407407407407, |
| "grad_norm": 1.515625, |
| "learning_rate": 1.9899098976621922e-05, |
| "loss": 0.8003520965576172, |
| "mean_token_accuracy": 0.7903904721140862, |
| "num_tokens": 9990163.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.7919679537415505, |
| "epoch": 0.7936507936507936, |
| "grad_norm": 1.8671875, |
| "learning_rate": 1.9873375335080915e-05, |
| "loss": 0.7965619087219238, |
| "mean_token_accuracy": 0.7903877586126328, |
| "num_tokens": 10710006.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.7846287094056607, |
| "epoch": 0.8465608465608465, |
| "grad_norm": 1.4453125, |
| "learning_rate": 1.9844754642743164e-05, |
| "loss": 0.7887547016143799, |
| "mean_token_accuracy": 0.7923174299299717, |
| "num_tokens": 11416822.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.787412092834711, |
| "epoch": 0.8994708994708994, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.981324529750672e-05, |
| "loss": 0.7970062732696533, |
| "mean_token_accuracy": 0.7903717122972012, |
| "num_tokens": 12123118.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.7589369006454945, |
| "epoch": 0.9523809523809523, |
| "grad_norm": 1.5, |
| "learning_rate": 1.9778856544859647e-05, |
| "loss": 0.7599987030029297, |
| "mean_token_accuracy": 0.7979414351284504, |
| "num_tokens": 12831058.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.7580704741179943, |
| "epoch": 1.0052910052910053, |
| "grad_norm": 1.3515625, |
| "learning_rate": 1.9741598475167173e-05, |
| "loss": 0.7588405609130859, |
| "mean_token_accuracy": 0.7996042884886265, |
| "num_tokens": 13547943.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.7310580350458622, |
| "epoch": 1.0582010582010581, |
| "grad_norm": 1.4765625, |
| "learning_rate": 1.9701482020711004e-05, |
| "loss": 0.7377434730529785, |
| "mean_token_accuracy": 0.8026577197015285, |
| "num_tokens": 14264772.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.7363193772733212, |
| "epoch": 1.1111111111111112, |
| "grad_norm": 1.3203125, |
| "learning_rate": 1.9658518952481543e-05, |
| "loss": 0.7403535842895508, |
| "mean_token_accuracy": 0.8024087257683277, |
| "num_tokens": 14985565.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.7300132818520069, |
| "epoch": 1.164021164021164, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.9612721876724054e-05, |
| "loss": 0.7335801124572754, |
| "mean_token_accuracy": 0.8039513885974884, |
| "num_tokens": 15701796.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.7273651890456676, |
| "epoch": 1.216931216931217, |
| "grad_norm": 1.40625, |
| "learning_rate": 1.9564104231239726e-05, |
| "loss": 0.7306139469146729, |
| "mean_token_accuracy": 0.8031473867595196, |
| "num_tokens": 16412522.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.7269332259893417, |
| "epoch": 1.2698412698412698, |
| "grad_norm": 1.5, |
| "learning_rate": 1.9512680281442745e-05, |
| "loss": 0.7315571308135986, |
| "mean_token_accuracy": 0.804153886437416, |
| "num_tokens": 17126523.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.7240817807614803, |
| "epoch": 1.3227513227513228, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.9458465116174534e-05, |
| "loss": 0.7265042781829834, |
| "mean_token_accuracy": 0.8053628757596016, |
| "num_tokens": 17834934.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.7229904316365718, |
| "epoch": 1.3756613756613756, |
| "grad_norm": 1.5078125, |
| "learning_rate": 1.940147464327637e-05, |
| "loss": 0.7304715156555176, |
| "mean_token_accuracy": 0.8046498209238052, |
| "num_tokens": 18553413.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.7136867262423039, |
| "epoch": 1.4285714285714286, |
| "grad_norm": 1.3984375, |
| "learning_rate": 1.9341725584921712e-05, |
| "loss": 0.7197102546691895, |
| "mean_token_accuracy": 0.8071116916835308, |
| "num_tokens": 19270191.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.7036095522344112, |
| "epoch": 1.4814814814814814, |
| "grad_norm": 1.375, |
| "learning_rate": 1.9279235472709554e-05, |
| "loss": 0.7028598785400391, |
| "mean_token_accuracy": 0.8090868189930915, |
| "num_tokens": 19976552.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.716102697700262, |
| "epoch": 1.5343915343915344, |
| "grad_norm": 1.359375, |
| "learning_rate": 1.92140226425203e-05, |
| "loss": 0.7210822105407715, |
| "mean_token_accuracy": 0.8061651162803173, |
| "num_tokens": 20676613.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.7108397655189037, |
| "epoch": 1.5873015873015874, |
| "grad_norm": 1.421875, |
| "learning_rate": 1.914610622913566e-05, |
| "loss": 0.7198187828063964, |
| "mean_token_accuracy": 0.8068007662892341, |
| "num_tokens": 21385978.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.7059788055717945, |
| "epoch": 1.6402116402116402, |
| "grad_norm": 1.4296875, |
| "learning_rate": 1.9075506160624068e-05, |
| "loss": 0.712452507019043, |
| "mean_token_accuracy": 0.8080652669072151, |
| "num_tokens": 22094336.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.6891738481819629, |
| "epoch": 1.693121693121693, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.900224315249343e-05, |
| "loss": 0.6894668579101563, |
| "mean_token_accuracy": 0.8126135811209678, |
| "num_tokens": 22798906.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.6981304191052914, |
| "epoch": 1.746031746031746, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.8926338701612738e-05, |
| "loss": 0.6995977401733399, |
| "mean_token_accuracy": 0.8106051124632359, |
| "num_tokens": 23511239.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.7145946308970451, |
| "epoch": 1.798941798941799, |
| "grad_norm": 1.53125, |
| "learning_rate": 1.884781507990445e-05, |
| "loss": 0.7227201461791992, |
| "mean_token_accuracy": 0.8075091950595379, |
| "num_tokens": 24231185.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.6977549403905868, |
| "epoch": 1.8518518518518519, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.8766695327809463e-05, |
| "loss": 0.7001574516296387, |
| "mean_token_accuracy": 0.8110232800245285, |
| "num_tokens": 24944285.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.6864141449332237, |
| "epoch": 1.9047619047619047, |
| "grad_norm": 1.4140625, |
| "learning_rate": 1.8683003247526567e-05, |
| "loss": 0.6919449806213379, |
| "mean_token_accuracy": 0.8123873800039292, |
| "num_tokens": 25670441.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.6988187648355961, |
| "epoch": 1.9576719576719577, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.8596763396028384e-05, |
| "loss": 0.7042366027832031, |
| "mean_token_accuracy": 0.8103295296430588, |
| "num_tokens": 26379786.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.6846350051462651, |
| "epoch": 2.0105820105820107, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.8508001077855876e-05, |
| "loss": 0.6828127861022949, |
| "mean_token_accuracy": 0.8147809013724328, |
| "num_tokens": 27097166.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.6520826548337937, |
| "epoch": 2.0634920634920633, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.8416742337693436e-05, |
| "loss": 0.6539791107177735, |
| "mean_token_accuracy": 0.8201325155794621, |
| "num_tokens": 27818991.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.6595748387277126, |
| "epoch": 2.1164021164021163, |
| "grad_norm": 1.515625, |
| "learning_rate": 1.8323013952726877e-05, |
| "loss": 0.6604544639587402, |
| "mean_token_accuracy": 0.8178910993039608, |
| "num_tokens": 28524822.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.6458595909178257, |
| "epoch": 2.1693121693121693, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.8226843424786434e-05, |
| "loss": 0.6502902984619141, |
| "mean_token_accuracy": 0.8199047416448593, |
| "num_tokens": 29245634.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.6555904000997543, |
| "epoch": 2.2222222222222223, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.8128258972277176e-05, |
| "loss": 0.6557448387145997, |
| "mean_token_accuracy": 0.8196609579026699, |
| "num_tokens": 29967135.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.6628798440098762, |
| "epoch": 2.2751322751322753, |
| "grad_norm": 1.4453125, |
| "learning_rate": 1.8027289521899143e-05, |
| "loss": 0.6660602569580079, |
| "mean_token_accuracy": 0.8170774318277836, |
| "num_tokens": 30680274.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.6558427393436432, |
| "epoch": 2.328042328042328, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.7923964700159658e-05, |
| "loss": 0.6595181465148926, |
| "mean_token_accuracy": 0.8196531683206558, |
| "num_tokens": 31386859.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.6561373583972454, |
| "epoch": 2.380952380952381, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.78183148246803e-05, |
| "loss": 0.658083152770996, |
| "mean_token_accuracy": 0.8189397811889648, |
| "num_tokens": 32111007.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.6530930824577809, |
| "epoch": 2.433862433862434, |
| "grad_norm": 1.4765625, |
| "learning_rate": 1.77103708953011e-05, |
| "loss": 0.656781816482544, |
| "mean_token_accuracy": 0.8194618344306945, |
| "num_tokens": 32819180.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.6579011715948582, |
| "epoch": 2.4867724867724865, |
| "grad_norm": 1.2578125, |
| "learning_rate": 1.7600164584984548e-05, |
| "loss": 0.6578071594238282, |
| "mean_token_accuracy": 0.8188325464725494, |
| "num_tokens": 33523113.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.6502429388463498, |
| "epoch": 2.5396825396825395, |
| "grad_norm": 1.3515625, |
| "learning_rate": 1.7487728230522085e-05, |
| "loss": 0.6535434722900391, |
| "mean_token_accuracy": 0.8208725921809673, |
| "num_tokens": 34240357.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.6465963438153267, |
| "epoch": 2.5925925925925926, |
| "grad_norm": 1.3515625, |
| "learning_rate": 1.737309482304585e-05, |
| "loss": 0.6498763084411621, |
| "mean_token_accuracy": 0.821516727656126, |
| "num_tokens": 34958032.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.6473522856831551, |
| "epoch": 2.6455026455026456, |
| "grad_norm": 1.3515625, |
| "learning_rate": 1.7256297998348388e-05, |
| "loss": 0.6505583763122559, |
| "mean_token_accuracy": 0.8211475618183612, |
| "num_tokens": 35658658.0, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.6565315008163453, |
| "epoch": 2.6984126984126986, |
| "grad_norm": 1.4140625, |
| "learning_rate": 1.7137372027013212e-05, |
| "loss": 0.6598291397094727, |
| "mean_token_accuracy": 0.8183600060641766, |
| "num_tokens": 36364754.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.6542541973292828, |
| "epoch": 2.751322751322751, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.7016351804359134e-05, |
| "loss": 0.6558931827545166, |
| "mean_token_accuracy": 0.820868706703186, |
| "num_tokens": 37074678.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.6618583984673023, |
| "epoch": 2.804232804232804, |
| "grad_norm": 1.4375, |
| "learning_rate": 1.6893272840201243e-05, |
| "loss": 0.6607325553894043, |
| "mean_token_accuracy": 0.8187017820775508, |
| "num_tokens": 37787128.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.6450109876692295, |
| "epoch": 2.857142857142857, |
| "grad_norm": 1.390625, |
| "learning_rate": 1.6768171248431603e-05, |
| "loss": 0.6493658065795899, |
| "mean_token_accuracy": 0.8208952791988849, |
| "num_tokens": 38504013.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.6399648249149322, |
| "epoch": 2.91005291005291, |
| "grad_norm": 1.4453125, |
| "learning_rate": 1.66410837364227e-05, |
| "loss": 0.6415593147277832, |
| "mean_token_accuracy": 0.8226414546370506, |
| "num_tokens": 39212782.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.6446338936686515, |
| "epoch": 2.962962962962963, |
| "grad_norm": 1.390625, |
| "learning_rate": 1.651204759425674e-05, |
| "loss": 0.6467597007751464, |
| "mean_token_accuracy": 0.822068239748478, |
| "num_tokens": 39931947.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.6519015111029148, |
| "epoch": 3.015873015873016, |
| "grad_norm": 1.65625, |
| "learning_rate": 1.6381100683783984e-05, |
| "loss": 0.6495416641235352, |
| "mean_token_accuracy": 0.8214401297271252, |
| "num_tokens": 40643225.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.6235358692705631, |
| "epoch": 3.068783068783069, |
| "grad_norm": 1.3203125, |
| "learning_rate": 1.6248281427513293e-05, |
| "loss": 0.6290417671203613, |
| "mean_token_accuracy": 0.8248508743941784, |
| "num_tokens": 41359577.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.6279302626848221, |
| "epoch": 3.121693121693122, |
| "grad_norm": 1.3828125, |
| "learning_rate": 1.6113628797338182e-05, |
| "loss": 0.62760910987854, |
| "mean_token_accuracy": 0.8258996732532978, |
| "num_tokens": 42068599.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.6285111606121063, |
| "epoch": 3.1746031746031744, |
| "grad_norm": 1.2578125, |
| "learning_rate": 1.597718230310167e-05, |
| "loss": 0.6302815437316894, |
| "mean_token_accuracy": 0.8256399400532246, |
| "num_tokens": 42785175.0, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.6026831187307835, |
| "epoch": 3.2275132275132274, |
| "grad_norm": 1.2421875, |
| "learning_rate": 1.5838981981003273e-05, |
| "loss": 0.6043402671813964, |
| "mean_token_accuracy": 0.8310895174741745, |
| "num_tokens": 43504295.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.6095856722444296, |
| "epoch": 3.2804232804232805, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.5699068381851576e-05, |
| "loss": 0.6121260643005371, |
| "mean_token_accuracy": 0.8288436524569989, |
| "num_tokens": 44210262.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.6268397398293019, |
| "epoch": 3.3333333333333335, |
| "grad_norm": 1.359375, |
| "learning_rate": 1.5557482559165777e-05, |
| "loss": 0.6273800373077393, |
| "mean_token_accuracy": 0.8257726937532425, |
| "num_tokens": 44921899.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.6130324799567461, |
| "epoch": 3.386243386243386, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.541426605712975e-05, |
| "loss": 0.6151331901550293, |
| "mean_token_accuracy": 0.8280435480177403, |
| "num_tokens": 45631735.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.6239284582436084, |
| "epoch": 3.439153439153439, |
| "grad_norm": 1.328125, |
| "learning_rate": 1.526946089840214e-05, |
| "loss": 0.6226998329162597, |
| "mean_token_accuracy": 0.8254561886191368, |
| "num_tokens": 46339604.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.6176637753844261, |
| "epoch": 3.492063492063492, |
| "grad_norm": 2.640625, |
| "learning_rate": 1.5123109571786034e-05, |
| "loss": 0.6186460971832275, |
| "mean_token_accuracy": 0.82830960303545, |
| "num_tokens": 47060193.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.6228313788771629, |
| "epoch": 3.544973544973545, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.4975255019761896e-05, |
| "loss": 0.6256577491760253, |
| "mean_token_accuracy": 0.8251520521938801, |
| "num_tokens": 47768090.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.6073577616363763, |
| "epoch": 3.597883597883598, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.4825940625887343e-05, |
| "loss": 0.608558177947998, |
| "mean_token_accuracy": 0.8303223170340062, |
| "num_tokens": 48486016.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.6105879176408052, |
| "epoch": 3.6507936507936507, |
| "grad_norm": 1.2578125, |
| "learning_rate": 1.467521020206753e-05, |
| "loss": 0.6111472129821778, |
| "mean_token_accuracy": 0.8288198597729206, |
| "num_tokens": 49196144.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.6225075088441372, |
| "epoch": 3.7037037037037037, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.4523107975699833e-05, |
| "loss": 0.6227302551269531, |
| "mean_token_accuracy": 0.8270458020269871, |
| "num_tokens": 49912021.0, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.6129852499812841, |
| "epoch": 3.7566137566137567, |
| "grad_norm": 1.21875, |
| "learning_rate": 1.4369678576696633e-05, |
| "loss": 0.616145133972168, |
| "mean_token_accuracy": 0.8275811091065407, |
| "num_tokens": 50634262.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.6109543219208717, |
| "epoch": 3.8095238095238093, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.4214967024389978e-05, |
| "loss": 0.6092634677886963, |
| "mean_token_accuracy": 0.8294319190084934, |
| "num_tokens": 51343452.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.6103949204087258, |
| "epoch": 3.8624338624338623, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.4059018714321986e-05, |
| "loss": 0.6095344066619873, |
| "mean_token_accuracy": 0.8291715182363987, |
| "num_tokens": 52046363.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.6032329283654689, |
| "epoch": 3.9153439153439153, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.390187940492486e-05, |
| "loss": 0.6063249588012696, |
| "mean_token_accuracy": 0.830570637434721, |
| "num_tokens": 52755135.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.5983160465955735, |
| "epoch": 3.9682539682539684, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.3743595204094442e-05, |
| "loss": 0.5990692138671875, |
| "mean_token_accuracy": 0.8316747449338436, |
| "num_tokens": 53473566.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.5978753294795751, |
| "epoch": 4.021164021164021, |
| "grad_norm": 1.203125, |
| "learning_rate": 1.3584212555661187e-05, |
| "loss": 0.5931243896484375, |
| "mean_token_accuracy": 0.833243765681982, |
| "num_tokens": 54190492.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.5927798122167587, |
| "epoch": 4.074074074074074, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.3423778225762609e-05, |
| "loss": 0.5956585884094239, |
| "mean_token_accuracy": 0.8326095908880233, |
| "num_tokens": 54897936.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.5964819733053446, |
| "epoch": 4.1269841269841265, |
| "grad_norm": 1.3203125, |
| "learning_rate": 1.3262339289121129e-05, |
| "loss": 0.5986158847808838, |
| "mean_token_accuracy": 0.8313788689672947, |
| "num_tokens": 55609388.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.5928852215409279, |
| "epoch": 4.1798941798941796, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.3099943115231386e-05, |
| "loss": 0.5939154148101806, |
| "mean_token_accuracy": 0.8326825730502605, |
| "num_tokens": 56329299.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.5908802170306444, |
| "epoch": 4.232804232804233, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.2936637354461064e-05, |
| "loss": 0.5890358924865723, |
| "mean_token_accuracy": 0.8333992123603821, |
| "num_tokens": 57042704.0, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.5940023452043534, |
| "epoch": 4.285714285714286, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.2772469924069285e-05, |
| "loss": 0.5959939002990723, |
| "mean_token_accuracy": 0.8325416699051857, |
| "num_tokens": 57755769.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.59011813364923, |
| "epoch": 4.338624338624339, |
| "grad_norm": 1.2421875, |
| "learning_rate": 1.2607488994146703e-05, |
| "loss": 0.5888036727905274, |
| "mean_token_accuracy": 0.83399997651577, |
| "num_tokens": 58467967.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.6021874897181988, |
| "epoch": 4.391534391534392, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.2441742973481432e-05, |
| "loss": 0.606382179260254, |
| "mean_token_accuracy": 0.8303363271057606, |
| "num_tokens": 59188579.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.5849491123110055, |
| "epoch": 4.444444444444445, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.2275280495354889e-05, |
| "loss": 0.5840800285339356, |
| "mean_token_accuracy": 0.8359167717397213, |
| "num_tokens": 59899589.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.5860148169100284, |
| "epoch": 4.497354497354498, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.21081504032718e-05, |
| "loss": 0.5868220329284668, |
| "mean_token_accuracy": 0.8345484808087349, |
| "num_tokens": 60620913.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.5894476883113384, |
| "epoch": 4.550264550264551, |
| "grad_norm": 1.3359375, |
| "learning_rate": 1.1940401736628524e-05, |
| "loss": 0.5873554229736329, |
| "mean_token_accuracy": 0.8340311080217362, |
| "num_tokens": 61330675.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.5952170636504889, |
| "epoch": 4.603174603174603, |
| "grad_norm": 1.3984375, |
| "learning_rate": 1.1772083716323878e-05, |
| "loss": 0.5960660934448242, |
| "mean_token_accuracy": 0.8311950400471687, |
| "num_tokens": 62041645.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.591047377511859, |
| "epoch": 4.656084656084656, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.1603245730316724e-05, |
| "loss": 0.5910706520080566, |
| "mean_token_accuracy": 0.8324602991342545, |
| "num_tokens": 62750446.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.583647707104683, |
| "epoch": 4.708994708994709, |
| "grad_norm": 1.359375, |
| "learning_rate": 1.1433937319134512e-05, |
| "loss": 0.5823469161987305, |
| "mean_token_accuracy": 0.835607772320509, |
| "num_tokens": 63460008.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.5815492067486048, |
| "epoch": 4.761904761904762, |
| "grad_norm": 1.3203125, |
| "learning_rate": 1.1264208161337091e-05, |
| "loss": 0.5845518112182617, |
| "mean_token_accuracy": 0.8349362075328827, |
| "num_tokens": 64164803.0, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.5895786888897419, |
| "epoch": 4.814814814814815, |
| "grad_norm": 1.3203125, |
| "learning_rate": 1.1094108058940005e-05, |
| "loss": 0.5885090827941895, |
| "mean_token_accuracy": 0.8332687266170978, |
| "num_tokens": 64878163.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.5861702430993319, |
| "epoch": 4.867724867724868, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.0923686922801537e-05, |
| "loss": 0.5883007049560547, |
| "mean_token_accuracy": 0.8344706572592259, |
| "num_tokens": 65599878.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.577086403593421, |
| "epoch": 4.920634920634921, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.0752994757977862e-05, |
| "loss": 0.575294017791748, |
| "mean_token_accuracy": 0.8364286012947559, |
| "num_tokens": 66312478.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.580839442834258, |
| "epoch": 4.973544973544973, |
| "grad_norm": 1.234375, |
| "learning_rate": 1.058208164905052e-05, |
| "loss": 0.5833872318267822, |
| "mean_token_accuracy": 0.8350156374275685, |
| "num_tokens": 67026652.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.57722695954144, |
| "epoch": 5.026455026455026, |
| "grad_norm": 1.3984375, |
| "learning_rate": 1.041099774543059e-05, |
| "loss": 0.5721875667572022, |
| "mean_token_accuracy": 0.837412741780281, |
| "num_tokens": 67743837.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.5689107570797205, |
| "epoch": 5.079365079365079, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.023979324664382e-05, |
| "loss": 0.5704532623291015, |
| "mean_token_accuracy": 0.8376903526484967, |
| "num_tokens": 68459077.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.566339960321784, |
| "epoch": 5.132275132275132, |
| "grad_norm": 1.3203125, |
| "learning_rate": 1.0068518387601038e-05, |
| "loss": 0.5641752243041992, |
| "mean_token_accuracy": 0.8387337259948253, |
| "num_tokens": 69175886.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.567414516210556, |
| "epoch": 5.185185185185185, |
| "grad_norm": 1.546875, |
| "learning_rate": 9.897223423858226e-06, |
| "loss": 0.5701710224151612, |
| "mean_token_accuracy": 0.8381049610674381, |
| "num_tokens": 69888377.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.5735208492726087, |
| "epoch": 5.238095238095238, |
| "grad_norm": 1.2578125, |
| "learning_rate": 9.725958616870495e-06, |
| "loss": 0.5715389728546143, |
| "mean_token_accuracy": 0.837762675434351, |
| "num_tokens": 70604584.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.5762054596096278, |
| "epoch": 5.291005291005291, |
| "grad_norm": 1.3046875, |
| "learning_rate": 9.554774219244344e-06, |
| "loss": 0.5787403106689453, |
| "mean_token_accuracy": 0.835259947925806, |
| "num_tokens": 71321169.0, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.5732409968972206, |
| "epoch": 5.343915343915344, |
| "grad_norm": 1.328125, |
| "learning_rate": 9.38372045999252e-06, |
| "loss": 0.5709231853485107, |
| "mean_token_accuracy": 0.8378458134829998, |
| "num_tokens": 72041175.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.5593106970191002, |
| "epoch": 5.396825396825397, |
| "grad_norm": 1.3359375, |
| "learning_rate": 9.212847529795784e-06, |
| "loss": 0.5596912384033204, |
| "mean_token_accuracy": 0.8397281520068646, |
| "num_tokens": 72749079.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.5708956997841597, |
| "epoch": 5.449735449735449, |
| "grad_norm": 1.1953125, |
| "learning_rate": 9.042205566275952e-06, |
| "loss": 0.5709175586700439, |
| "mean_token_accuracy": 0.8381867438554764, |
| "num_tokens": 73462893.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.574559985101223, |
| "epoch": 5.502645502645502, |
| "grad_norm": 1.25, |
| "learning_rate": 8.871844639284479e-06, |
| "loss": 0.5761849403381347, |
| "mean_token_accuracy": 0.8366024740040302, |
| "num_tokens": 74167133.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.5802072331309318, |
| "epoch": 5.555555555555555, |
| "grad_norm": 1.1875, |
| "learning_rate": 8.701814736210933e-06, |
| "loss": 0.5792787075042725, |
| "mean_token_accuracy": 0.835508918017149, |
| "num_tokens": 74870399.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.5814166940748692, |
| "epoch": 5.608465608465608, |
| "grad_norm": 1.28125, |
| "learning_rate": 8.532165747315686e-06, |
| "loss": 0.5834609031677246, |
| "mean_token_accuracy": 0.8350230395793915, |
| "num_tokens": 75590782.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.5799121633172035, |
| "epoch": 5.661375661375661, |
| "grad_norm": 1.3203125, |
| "learning_rate": 8.362947451091072e-06, |
| "loss": 0.5762242317199707, |
| "mean_token_accuracy": 0.8368450194597244, |
| "num_tokens": 76305788.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.5694117113947869, |
| "epoch": 5.714285714285714, |
| "grad_norm": 1.25, |
| "learning_rate": 8.194209499655364e-06, |
| "loss": 0.5744447231292724, |
| "mean_token_accuracy": 0.8372330017387867, |
| "num_tokens": 77023579.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.5791369702666997, |
| "epoch": 5.767195767195767, |
| "grad_norm": 1.34375, |
| "learning_rate": 8.026001404183831e-06, |
| "loss": 0.574954605102539, |
| "mean_token_accuracy": 0.8362665802240372, |
| "num_tokens": 77720821.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.575787166133523, |
| "epoch": 5.8201058201058204, |
| "grad_norm": 1.2265625, |
| "learning_rate": 7.85837252038112e-06, |
| "loss": 0.5779669761657715, |
| "mean_token_accuracy": 0.83624067902565, |
| "num_tokens": 78431671.0, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.5742827598005533, |
| "epoch": 5.8730158730158735, |
| "grad_norm": 1.359375, |
| "learning_rate": 7.69137203399931e-06, |
| "loss": 0.5727415084838867, |
| "mean_token_accuracy": 0.8372173503041267, |
| "num_tokens": 79141393.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.561671144515276, |
| "epoch": 5.925925925925926, |
| "grad_norm": 1.2578125, |
| "learning_rate": 7.525048946405791e-06, |
| "loss": 0.5620074272155762, |
| "mean_token_accuracy": 0.8406617894768715, |
| "num_tokens": 79858356.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.5741465941071511, |
| "epoch": 5.978835978835979, |
| "grad_norm": 1.2890625, |
| "learning_rate": 7.359452060205245e-06, |
| "loss": 0.5751906394958496, |
| "mean_token_accuracy": 0.8366710051894188, |
| "num_tokens": 80566532.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.5611932255327702, |
| "epoch": 6.031746031746032, |
| "grad_norm": 1.2109375, |
| "learning_rate": 7.194629964919989e-06, |
| "loss": 0.5557641983032227, |
| "mean_token_accuracy": 0.8405756376683712, |
| "num_tokens": 81283415.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.5589057210832834, |
| "epoch": 6.084656084656085, |
| "grad_norm": 1.265625, |
| "learning_rate": 7.030631022732781e-06, |
| "loss": 0.5576947212219239, |
| "mean_token_accuracy": 0.8407048217952251, |
| "num_tokens": 81995607.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.565440958738327, |
| "epoch": 6.137566137566138, |
| "grad_norm": 1.3359375, |
| "learning_rate": 6.867503354296392e-06, |
| "loss": 0.5673689842224121, |
| "mean_token_accuracy": 0.8382438689470291, |
| "num_tokens": 82712148.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.5743044268339872, |
| "epoch": 6.190476190476191, |
| "grad_norm": 1.28125, |
| "learning_rate": 6.705294824614004e-06, |
| "loss": 0.570713996887207, |
| "mean_token_accuracy": 0.8369224861264228, |
| "num_tokens": 83425506.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.564292473345995, |
| "epoch": 6.243386243386244, |
| "grad_norm": 1.34375, |
| "learning_rate": 6.54405302899463e-06, |
| "loss": 0.564542007446289, |
| "mean_token_accuracy": 0.8393503338098526, |
| "num_tokens": 84137018.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.5736533626914024, |
| "epoch": 6.296296296296296, |
| "grad_norm": 1.359375, |
| "learning_rate": 6.383825279087699e-06, |
| "loss": 0.5708751678466797, |
| "mean_token_accuracy": 0.8368868716061115, |
| "num_tokens": 84849598.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.5581273920834064, |
| "epoch": 6.349206349206349, |
| "grad_norm": 1.265625, |
| "learning_rate": 6.2246585890008245e-06, |
| "loss": 0.5611091613769531, |
| "mean_token_accuracy": 0.8401936620473862, |
| "num_tokens": 85561962.0, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.5584203939884901, |
| "epoch": 6.402116402116402, |
| "grad_norm": 1.2890625, |
| "learning_rate": 6.06659966150491e-06, |
| "loss": 0.5574289798736572, |
| "mean_token_accuracy": 0.8403389886021614, |
| "num_tokens": 86286338.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.5638503111898899, |
| "epoch": 6.455026455026455, |
| "grad_norm": 1.2890625, |
| "learning_rate": 5.909694874330601e-06, |
| "loss": 0.5615947246551514, |
| "mean_token_accuracy": 0.8403584957122803, |
| "num_tokens": 87001911.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.5603939324617386, |
| "epoch": 6.507936507936508, |
| "grad_norm": 1.2421875, |
| "learning_rate": 5.753990266560105e-06, |
| "loss": 0.5629959106445312, |
| "mean_token_accuracy": 0.840188529342413, |
| "num_tokens": 87714571.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.56005436219275, |
| "epoch": 6.560846560846561, |
| "grad_norm": 1.296875, |
| "learning_rate": 5.599531525118373e-06, |
| "loss": 0.5609263420104981, |
| "mean_token_accuracy": 0.8397768594324588, |
| "num_tokens": 88417674.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.5575656868517399, |
| "epoch": 6.613756613756614, |
| "grad_norm": 1.2734375, |
| "learning_rate": 5.446363971367638e-06, |
| "loss": 0.5569974899291992, |
| "mean_token_accuracy": 0.8405981361865997, |
| "num_tokens": 89123522.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.5667697321623564, |
| "epoch": 6.666666666666667, |
| "grad_norm": 1.234375, |
| "learning_rate": 5.29453254780917e-06, |
| "loss": 0.5631841659545899, |
| "mean_token_accuracy": 0.8394736826419831, |
| "num_tokens": 89838831.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.5579639460891486, |
| "epoch": 6.71957671957672, |
| "grad_norm": 1.203125, |
| "learning_rate": 5.1440818048962575e-06, |
| "loss": 0.557750654220581, |
| "mean_token_accuracy": 0.8407927252352237, |
| "num_tokens": 90546066.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.5646323654800653, |
| "epoch": 6.772486772486772, |
| "grad_norm": 1.21875, |
| "learning_rate": 4.995055887962167e-06, |
| "loss": 0.5658845901489258, |
| "mean_token_accuracy": 0.8390677250921726, |
| "num_tokens": 91257301.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.5639095064252615, |
| "epoch": 6.825396825396825, |
| "grad_norm": 1.3203125, |
| "learning_rate": 4.8474985242669965e-06, |
| "loss": 0.5612518310546875, |
| "mean_token_accuracy": 0.840159710496664, |
| "num_tokens": 91961075.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.5754041139036417, |
| "epoch": 6.878306878306878, |
| "grad_norm": 1.375, |
| "learning_rate": 4.701453010167217e-06, |
| "loss": 0.5760274887084961, |
| "mean_token_accuracy": 0.8366920158267022, |
| "num_tokens": 92677769.0, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.5579227816313506, |
| "epoch": 6.931216931216931, |
| "grad_norm": 1.265625, |
| "learning_rate": 4.556962198411631e-06, |
| "loss": 0.5589898109436036, |
| "mean_token_accuracy": 0.8401356041431427, |
| "num_tokens": 93390418.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.5532530311495065, |
| "epoch": 6.984126984126984, |
| "grad_norm": 1.3359375, |
| "learning_rate": 4.414068485567473e-06, |
| "loss": 0.5527856826782227, |
| "mean_token_accuracy": 0.8411326013505459, |
| "num_tokens": 94107638.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.5631154362112284, |
| "epoch": 7.037037037037037, |
| "grad_norm": 1.234375, |
| "learning_rate": 4.272813799580417e-06, |
| "loss": 0.5605039119720459, |
| "mean_token_accuracy": 0.8398166857659817, |
| "num_tokens": 94829458.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.5513886582106352, |
| "epoch": 7.08994708994709, |
| "grad_norm": 1.2265625, |
| "learning_rate": 4.1332395874720474e-06, |
| "loss": 0.5492176532745361, |
| "mean_token_accuracy": 0.8416748136281967, |
| "num_tokens": 95540140.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.558628436177969, |
| "epoch": 7.142857142857143, |
| "grad_norm": 1.2890625, |
| "learning_rate": 3.9953868031784415e-06, |
| "loss": 0.5589038848876953, |
| "mean_token_accuracy": 0.8402540810406208, |
| "num_tokens": 96254434.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.570279137045145, |
| "epoch": 7.195767195767195, |
| "grad_norm": 1.3984375, |
| "learning_rate": 3.8592958955334636e-06, |
| "loss": 0.5722630023956299, |
| "mean_token_accuracy": 0.8379042416810989, |
| "num_tokens": 96966014.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.5559218022972345, |
| "epoch": 7.248677248677248, |
| "grad_norm": 1.28125, |
| "learning_rate": 3.725006796400231e-06, |
| "loss": 0.5557093620300293, |
| "mean_token_accuracy": 0.8406034111976624, |
| "num_tokens": 97677188.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.567194889113307, |
| "epoch": 7.301587301587301, |
| "grad_norm": 1.3515625, |
| "learning_rate": 3.5925589089542954e-06, |
| "loss": 0.5640733242034912, |
| "mean_token_accuracy": 0.8379683151841164, |
| "num_tokens": 98391677.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.5579260598868132, |
| "epoch": 7.354497354497354, |
| "grad_norm": 1.2734375, |
| "learning_rate": 3.461991096121955e-06, |
| "loss": 0.5559053421020508, |
| "mean_token_accuracy": 0.8408147536218167, |
| "num_tokens": 99105789.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.560818263143301, |
| "epoch": 7.407407407407407, |
| "grad_norm": 1.25, |
| "learning_rate": 3.3333416691770402e-06, |
| "loss": 0.5586414337158203, |
| "mean_token_accuracy": 0.8399199068546295, |
| "num_tokens": 99819774.0, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.5562900472432375, |
| "epoch": 7.4603174603174605, |
| "grad_norm": 1.28125, |
| "learning_rate": 3.206648376499637e-06, |
| "loss": 0.556308650970459, |
| "mean_token_accuracy": 0.8412726759910584, |
| "num_tokens": 100540345.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.5577713888138532, |
| "epoch": 7.5132275132275135, |
| "grad_norm": 1.28125, |
| "learning_rate": 3.0819483924999183e-06, |
| "loss": 0.5583217144012451, |
| "mean_token_accuracy": 0.840390095859766, |
| "num_tokens": 101261007.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.5589878562837839, |
| "epoch": 7.5661375661375665, |
| "grad_norm": 1.2890625, |
| "learning_rate": 2.959278306710409e-06, |
| "loss": 0.558604097366333, |
| "mean_token_accuracy": 0.8404027812182904, |
| "num_tokens": 101972692.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.5598054159432649, |
| "epoch": 7.619047619047619, |
| "grad_norm": 1.21875, |
| "learning_rate": 2.838674113049884e-06, |
| "loss": 0.5580643653869629, |
| "mean_token_accuracy": 0.8400710172951221, |
| "num_tokens": 102684836.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.57165152952075, |
| "epoch": 7.671957671957672, |
| "grad_norm": 1.3828125, |
| "learning_rate": 2.7201711992619873e-06, |
| "loss": 0.5705924034118652, |
| "mean_token_accuracy": 0.8384876281023026, |
| "num_tokens": 103404477.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.5632453311234713, |
| "epoch": 7.724867724867725, |
| "grad_norm": 1.328125, |
| "learning_rate": 2.6038043365317643e-06, |
| "loss": 0.5622864246368409, |
| "mean_token_accuracy": 0.8403951473534107, |
| "num_tokens": 104111178.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.5514033962041139, |
| "epoch": 7.777777777777778, |
| "grad_norm": 1.25, |
| "learning_rate": 2.489607669283088e-06, |
| "loss": 0.5483118057250976, |
| "mean_token_accuracy": 0.8429030381143093, |
| "num_tokens": 104820924.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.5669628646224737, |
| "epoch": 7.830687830687831, |
| "grad_norm": 1.28125, |
| "learning_rate": 2.3776147051599518e-06, |
| "loss": 0.5667363166809082, |
| "mean_token_accuracy": 0.838956668972969, |
| "num_tokens": 105525661.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.5724949978291989, |
| "epoch": 7.883597883597884, |
| "grad_norm": 1.2734375, |
| "learning_rate": 2.2678583051946733e-06, |
| "loss": 0.5757796287536621, |
| "mean_token_accuracy": 0.8371090695261956, |
| "num_tokens": 106244487.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.5733943063765764, |
| "epoch": 7.936507936507937, |
| "grad_norm": 1.3359375, |
| "learning_rate": 2.1603706741657803e-06, |
| "loss": 0.572456169128418, |
| "mean_token_accuracy": 0.8370349399745465, |
| "num_tokens": 106950025.0, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.5639591090381145, |
| "epoch": 7.98941798941799, |
| "grad_norm": 1.2578125, |
| "learning_rate": 2.055183351148463e-06, |
| "loss": 0.5623289585113526, |
| "mean_token_accuracy": 0.8393302865326404, |
| "num_tokens": 107659233.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.5514031689614057, |
| "epoch": 8.042328042328043, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.9523272002603745e-06, |
| "loss": 0.5499680042266846, |
| "mean_token_accuracy": 0.8424346372485161, |
| "num_tokens": 108374173.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.5448824431747198, |
| "epoch": 8.095238095238095, |
| "grad_norm": 1.203125, |
| "learning_rate": 1.8518324016054601e-06, |
| "loss": 0.5410975456237793, |
| "mean_token_accuracy": 0.8450842514634133, |
| "num_tokens": 109085583.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.5542028229683638, |
| "epoch": 8.148148148148149, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.7537284424184852e-06, |
| "loss": 0.5530242919921875, |
| "mean_token_accuracy": 0.8411494240164756, |
| "num_tokens": 109801232.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.5648227866739035, |
| "epoch": 8.201058201058201, |
| "grad_norm": 1.3046875, |
| "learning_rate": 1.658044108412905e-06, |
| "loss": 0.5639634132385254, |
| "mean_token_accuracy": 0.8391084566712379, |
| "num_tokens": 110516150.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.5628920037299394, |
| "epoch": 8.253968253968253, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.564807475334509e-06, |
| "loss": 0.5674610137939453, |
| "mean_token_accuracy": 0.8390280671417714, |
| "num_tokens": 111232078.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.5625626020133495, |
| "epoch": 8.306878306878307, |
| "grad_norm": 1.5, |
| "learning_rate": 1.4740459007234498e-06, |
| "loss": 0.566582441329956, |
| "mean_token_accuracy": 0.8388344503939151, |
| "num_tokens": 111950711.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.5617518879473209, |
| "epoch": 8.359788359788359, |
| "grad_norm": 1.25, |
| "learning_rate": 1.3857860158869696e-06, |
| "loss": 0.5623356342315674, |
| "mean_token_accuracy": 0.8393992736935616, |
| "num_tokens": 112658277.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.5684962764382362, |
| "epoch": 8.412698412698413, |
| "grad_norm": 1.2265625, |
| "learning_rate": 1.3000537180852213e-06, |
| "loss": 0.5678952217102051, |
| "mean_token_accuracy": 0.8387253440916538, |
| "num_tokens": 113378303.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.5580512210726738, |
| "epoch": 8.465608465608465, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.2168741629324944e-06, |
| "loss": 0.5555991172790528, |
| "mean_token_accuracy": 0.8412777297198772, |
| "num_tokens": 114087597.0, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.5545306473970413, |
| "epoch": 8.518518518518519, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.1362717570160387e-06, |
| "loss": 0.5534980297088623, |
| "mean_token_accuracy": 0.8419035091996193, |
| "num_tokens": 114803563.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.5478607811033726, |
| "epoch": 8.571428571428571, |
| "grad_norm": 1.1796875, |
| "learning_rate": 1.0582701507346582e-06, |
| "loss": 0.5463678359985351, |
| "mean_token_accuracy": 0.8439272284507752, |
| "num_tokens": 115519221.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.5581431291997433, |
| "epoch": 8.624338624338625, |
| "grad_norm": 1.2734375, |
| "learning_rate": 9.828922313592227e-07, |
| "loss": 0.555583906173706, |
| "mean_token_accuracy": 0.8411937601864338, |
| "num_tokens": 116238487.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.5540006071329117, |
| "epoch": 8.677248677248677, |
| "grad_norm": 1.28125, |
| "learning_rate": 9.10160116317047e-07, |
| "loss": 0.5512968063354492, |
| "mean_token_accuracy": 0.8420908160507679, |
| "num_tokens": 116956616.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.5638836394995451, |
| "epoch": 8.73015873015873, |
| "grad_norm": 1.28125, |
| "learning_rate": 8.400951467022067e-07, |
| "loss": 0.5646930694580078, |
| "mean_token_accuracy": 0.8391482375562191, |
| "num_tokens": 117662184.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.5587781198322773, |
| "epoch": 8.783068783068783, |
| "grad_norm": 1.28125, |
| "learning_rate": 7.727178810136094e-07, |
| "loss": 0.5597066879272461, |
| "mean_token_accuracy": 0.8408370740711689, |
| "num_tokens": 118365440.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.5502323783934117, |
| "epoch": 8.835978835978835, |
| "grad_norm": 1.25, |
| "learning_rate": 7.080480891227126e-07, |
| "loss": 0.5470727920532227, |
| "mean_token_accuracy": 0.843222300708294, |
| "num_tokens": 119078464.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.5644949089735747, |
| "epoch": 8.88888888888889, |
| "grad_norm": 1.1953125, |
| "learning_rate": 6.461047464726433e-07, |
| "loss": 0.5681308746337891, |
| "mean_token_accuracy": 0.8390718877315522, |
| "num_tokens": 119792226.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.5596996039152146, |
| "epoch": 8.941798941798941, |
| "grad_norm": 1.2890625, |
| "learning_rate": 5.869060285104089e-07, |
| "loss": 0.5615987300872802, |
| "mean_token_accuracy": 0.8403952471911907, |
| "num_tokens": 120500468.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.5492101375013589, |
| "epoch": 8.994708994708995, |
| "grad_norm": 1.25, |
| "learning_rate": 5.304693053538434e-07, |
| "loss": 0.5487012386322021, |
| "mean_token_accuracy": 0.8426029972732068, |
| "num_tokens": 121205368.0, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.5425628896802663, |
| "epoch": 9.047619047619047, |
| "grad_norm": 1.2890625, |
| "learning_rate": 4.768111366948736e-07, |
| "loss": 0.542825984954834, |
| "mean_token_accuracy": 0.8438372492790223, |
| "num_tokens": 121917493.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.5565537832677364, |
| "epoch": 9.100529100529101, |
| "grad_norm": 1.3125, |
| "learning_rate": 4.2594726694055397e-07, |
| "loss": 0.5565238952636719, |
| "mean_token_accuracy": 0.8415951780974865, |
| "num_tokens": 122643710.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.5517397329211235, |
| "epoch": 9.153439153439153, |
| "grad_norm": 1.265625, |
| "learning_rate": 3.778926205933342e-07, |
| "loss": 0.5498518466949462, |
| "mean_token_accuracy": 0.8425840228796005, |
| "num_tokens": 123357455.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.5513115633279085, |
| "epoch": 9.206349206349206, |
| "grad_norm": 1.234375, |
| "learning_rate": 3.3266129787191015e-07, |
| "loss": 0.5506171226501465, |
| "mean_token_accuracy": 0.8420891851186753, |
| "num_tokens": 124071350.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.5496289785951376, |
| "epoch": 9.25925925925926, |
| "grad_norm": 1.28125, |
| "learning_rate": 2.9026657057392715e-07, |
| "loss": 0.5469128608703613, |
| "mean_token_accuracy": 0.8435498252511024, |
| "num_tokens": 124788814.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.5514781590551138, |
| "epoch": 9.312169312169312, |
| "grad_norm": 1.28125, |
| "learning_rate": 2.507208781817638e-07, |
| "loss": 0.5516631126403808, |
| "mean_token_accuracy": 0.841703935712576, |
| "num_tokens": 125504162.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.5572028573602438, |
| "epoch": 9.365079365079366, |
| "grad_norm": 1.359375, |
| "learning_rate": 2.1403582421253398e-07, |
| "loss": 0.5524599075317382, |
| "mean_token_accuracy": 0.841224092245102, |
| "num_tokens": 126213099.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.5627590391784907, |
| "epoch": 9.417989417989418, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.8022217281336908e-07, |
| "loss": 0.561534309387207, |
| "mean_token_accuracy": 0.8392536200582981, |
| "num_tokens": 126934123.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.5527021780610084, |
| "epoch": 9.470899470899472, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.4928984560301006e-07, |
| "loss": 0.550753116607666, |
| "mean_token_accuracy": 0.8426078997552395, |
| "num_tokens": 127647228.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.5678260687738657, |
| "epoch": 9.523809523809524, |
| "grad_norm": 1.375, |
| "learning_rate": 1.212479187605897e-07, |
| "loss": 0.5664829730987548, |
| "mean_token_accuracy": 0.8389328278601169, |
| "num_tokens": 128350815.0, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.5642268594354392, |
| "epoch": 9.576719576719576, |
| "grad_norm": 1.265625, |
| "learning_rate": 9.610462036249823e-08, |
| "loss": 0.5651995182037354, |
| "mean_token_accuracy": 0.8384916484355927, |
| "num_tokens": 129070776.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.5517740231007338, |
| "epoch": 9.62962962962963, |
| "grad_norm": 1.2734375, |
| "learning_rate": 7.386732796810014e-08, |
| "loss": 0.5510141372680664, |
| "mean_token_accuracy": 0.8419340692460537, |
| "num_tokens": 129784905.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.5449846308678389, |
| "epoch": 9.682539682539682, |
| "grad_norm": 1.2890625, |
| "learning_rate": 5.454256645500477e-08, |
| "loss": 0.5425686836242676, |
| "mean_token_accuracy": 0.8441961273550987, |
| "num_tokens": 130492958.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.5615681868046523, |
| "epoch": 9.735449735449736, |
| "grad_norm": 1.375, |
| "learning_rate": 3.8136006104533276e-08, |
| "loss": 0.5602152824401856, |
| "mean_token_accuracy": 0.840794712305069, |
| "num_tokens": 131214260.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.5509412154555321, |
| "epoch": 9.788359788359788, |
| "grad_norm": 1.359375, |
| "learning_rate": 2.4652460937941714e-08, |
| "loss": 0.5517104148864747, |
| "mean_token_accuracy": 0.8420154705643654, |
| "num_tokens": 131927722.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.5618329748511315, |
| "epoch": 9.841269841269842, |
| "grad_norm": 1.2890625, |
| "learning_rate": 1.4095887303888778e-08, |
| "loss": 0.5613327026367188, |
| "mean_token_accuracy": 0.839391653239727, |
| "num_tokens": 132622563.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.5594605170190334, |
| "epoch": 9.894179894179894, |
| "grad_norm": 1.2421875, |
| "learning_rate": 6.469382717563255e-09, |
| "loss": 0.5612988471984863, |
| "mean_token_accuracy": 0.8398191019892692, |
| "num_tokens": 133332199.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.5600211407989264, |
| "epoch": 9.947089947089948, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.775184951806619e-09, |
| "loss": 0.5590639114379883, |
| "mean_token_accuracy": 0.8405914202332496, |
| "num_tokens": 134041278.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.5648111406713724, |
| "epoch": 10.0, |
| "grad_norm": 1.296875, |
| "learning_rate": 1.4671380509367893e-11, |
| "loss": 0.5674919128417969, |
| "mean_token_accuracy": 0.8386900670826435, |
| "num_tokens": 134755992.0, |
| "step": 1890 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 1890, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.4340057784875745e+18, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
